State ofthe Art in Digital Pathology

Part II

State of the Art in Digital Pathology

Perspectives on Digital Pathology M. García-Rojo et al. (Eds.)

15

IOS Press, 2012

© 20J2 The aiithors and IOS Press. All rights reserx'ed. doi: 10.3233/978-1-61499-086-4-15

State of the Art and Trends for Digital Pathology Marcial GARCÍA ROJO^

Pathology Department, Hospital General Universitario de Ciudad Real, Spain; Chair, COST Action 1C0604 Euro-Telepath

Abstract. Anatomic pathology is a medical specialty where both information management systems and digital images systems paly a most important role. Digital pathology is a new concept that considers all uses of this information,

including diagnosis, biomedical research and education. Virtual microscopy or whole slide imaging, resulting in digital slides, is an outreaching technology in anatomic pathology. Limiting factors in the expansión of virtual microscopy are fonnidable storage dimensión, scanning speed, quality of image and cultural change. Anatomic pathology data and images should be an important part of the patient electronic health records as well as of clinical data warehouse, epidemiological or biomedical research databases, and platforms dedicated to translational medicine. Integrating anatomic pathology to the "healthcare enterprise" can only be achieved using existing and emerging medical informatics standards like Digital hnaging and Communications in Medicine (DICOM®'), Health Level Seven (HL7®), and Systematized Nomenclature of Medicine-

Clinical Terms (SNOMED CT®), following the recommendations of hitegrating the Healthcare Enterprise (IHE®). The consequences of the fiill digitalization of pathology departments are hard to foresee, but short term issues have arisen that imply interesting challenges for health care standards bodies Keywords. Digital pathology, healthcare, research, education, information systems, e-Health record, telepathology, virtual microscopy.

Introduction

Digital pathology concept comprises the information technology that allows for the management of information, including data and images, generated in an anatomic pathology department. Anatomic pathology information system and digital imaging modalities are two main components of digital pathology. Anatomic pathology information systems manage computerized orders of anatomic pathology examination of specimen collected from patients as well as the fiilfillment of these orders.

In most anatomic pathology departments, images are more and more being used in a digital format. Photographic images are digitized during specimen macroscopic study and microscopic images during microscopy evaluation and molecular pathology documentation.

' Corresponding Author. Marcial García Rojo, MD, PhD; Dep. of Pathology, Hospital General Universitario de Ciudad Real, Calle Obispo Rafael Torija s/n, E-13005 Ciudad Real, Spain; Email: [email protected] /eurotelepath/

Phone:

+34-926-278166;

Fax:

+34-926-278186;

URL:

http://www.conganat.org

16

M. García Rojo /State ofthe Art and Trendsfor Digital Pathology

Nowadays, most of the anatomic pathology departments using digital imaging in microscopy, which is the essential tool of pathologists, are equipped with imaging modalities producing only still images captured from a selected field under the microscope. The use of still microscopio images in the diagnostic process is time consuming and results in frustration for clinicians and pathologists. In the last 5 years, slide scanners have become very popular since they allow a complete digitization of tissue and cytology slides, a process ternied whole slides imaging (WSI), to create digital slides. Since this disruptive innovation allows a realistic simulation of the work performed with the conventional optical microscope (Figure 1), it is known as virtual microscopy [1,2].

Conventional glass slides are fragile, and they are non-permanent, since stain will fade over time, especially in immunofluorescence. Also, in cytology, it is not possible to distribute copies of the slides. These disadvantages of conventional slides can be overcome with digital slides, which also have additional advantages, like having lower magnification pictures with extraordinary quality, a dynamic map of the slide, tracking and playing back the path followed by the pathologist during the examination of the slide, images are permanently stored, and it is possible to make annotations that can also be recorded. Additional advantages of digital slides over glass slides are the possibility to distribute múltiple copies on a DVD of exactly the same slide, and the high quality always on focus and well illuminated images that decreases eye fatigue, and other ergonomic advantages by reducing back and neck fatigue [3]. Virtual microscopy is already being widely applied in anatomic pathology for primary histopathology diagnosis, primary cytopathology diagnosis, cytological cáncer screening. Recently the use of digital images is been validated for diagnostic applications in surgical pathology [4], cytopathology [5], and immunohistochemistry [6]. There is no statistically significant difference in the diagnostic accuracy either between virtual microscopy and conventional light microscopy. Some automated image analysis algorithms that are being used on digital slides have been U.S. Food and Drug Administration (FDA) 510(k)-cleared for assessing the level or certain immunohistochemical markers. However, there is little experience in the use of virtual microscopy in diagnostic pathology daily practice, and there in no slide scanner that is FDA approved for primary or initial diagnosis.

Digital pathology in an anatomic pathology department also includes biomedical equipment, like laboratory autostainers control software, automated image analysis tools, quality assurance program management systems, etc., that are used during the ftilfillment of the orders. This equipment needs to be properly interfaced to the anatomic pathology information system.

Digital pathology also addresses information technology used for inter-laboratories exchanges for collaborative diagnostic processes including telepathology for second opinion.

Telepathology is the practice of pathology at a distance using some ofthe elements that compose digital pathology (still images, digital slides, and less frequently video microscopy). It can be aimed to primary diagnosis or to second opinion teleconsultation [7].

Telepathology developed in some countries as a response to a shortage of pathologists [8]. It was classically divided into static telepathology (store and forward, generally sending still images files of representative areas by e-mail or FTP) and dynamic or real time telepathology (video transmission) [9]. Digital or virtual slides can be considered a special type of store and forward telepathology where the whole

M. García Rojo / State of the Art and Trendsfor Digital Pathology

17

slide is scaimed and saved, to be transmitted under user-demand, with the advantage of sending only requested areas that can be sent using conventional Internet bandwidth. Due to this dynamic interaction between users and the digital slide, some users classify virtual microscopy as digital dynamic telepathology systems [2]. In Japan, since 2000, telepathology is included as an insured healthcare service [8]. At last, digital pathology is not only dedicated to patient care and also addresses the use of information systems during research and teaching activities. Education librarles, biobanks managing systems, or cáncer registry databases are part of digital pathology. Virtual microscopy, for example, is already being widely applied in undergraduate teaching, distance learning and continuous medical education, proficiency testing, pathology quality assurance programs, and research (tumor banking). Digital pathology comprises the use of information technology for all the different activities of the pathologists and should be integrated to the "Healthcare Enterprise" information system for patient care (integration to electronic healthcare record or personal healthcare record) and research activities (integration to biomedical or epidemiological research databases, to cáncer registries, to tissue or images banks, to clinical data warehouse, to translational medicine platforms, etc.).

1. Electronic Pathology Study Orders

Typical anatomic pathology studies represent the examination of tissue removed in a single collection procedure (e.g., surgical operation/event, biopsy, scrape, aspiration etc.). In diagnostic activities, digital pathology addresses the management of both digital (or computerized) orders for anatomic pathology examination of specimen and digital reports, delivered in fulfillment of these orders. The electronic request from the electronic medical record (EMR), to facilítate patient identification and specimen registration in the pathology department, is of paramount importance. In anatomic pathology, the diagnostic process in anatomic pathology is specimendriven. This process differs from that in the clinical laboratory since it relies on image interpretation. It also differs from that in radiology since - in the current daily practice digital imaging is not systematically performed. Both in hospitals (outpatient clinic, hospital wards, and surgery rooms) and in primary care centers, integration with the electronic health record with the pathology information system should include features such as: • Electronic requests of biopsy, cytology, autopsy, and molecular pathology; • Fine needle aspiration (FNA) clinic, where this procedure if performed by the pathologist and other procedures (fresh uriñe collection) that may need an appointment; • Preparation of containers and fixatives; • Transportation of fresh and fixed specimens; • Report and pathology images searches (preferably through a web client).

18

M. García Rojo / State ofthe Art and Trendsfor Digital Pathology

2. Anatomic Pathology Information System

The main objective of the pathology information system is managing data and images efficiently to generate pathology reports, which are incorporated into the patient's medical record, but it also has other secondary uses (statistics, correlation cytohistological, tumor registry, and biobanks). For this it is necessary to have tools that allow data clustering or statistics that help in decision-making. Information systems based on web environment seem to be the best option nowadays, in order to facilítate equipment installation and management of versions and updates, among other advantages. Currently, in Spain, there are six commercial pathology information systems available, although only two of them are available as web clients (no installation required on each computer). The College of American Pathologists includes on its website information on 24 products for Pathology, of which 7 (30%) are available with thin web clients [10]. The specific functions of the anatomic pathology information system (APIS) are summarized as follows:

• •

Identification and management of patients. Registration and specimen management (biopsy, cytology, autopsy, molecular

•

pathology). Reports editing and workflow (clinical,

macroscopic and microscopic descriptions, final diagnosis, conclusions, referral to other information systems).

• • •

Archiving of direct specimens (tissue and liquid-based cytology), processed specimens, paraffin blocks and preparations. Macroscopic study. This is the gross examination of biopsies or autopsies. Traceability system and workflow control in laboratories. Generally, it manages the work of the pathology technicians and biologists. It is subdivided into

o o o o o

•

•

Biopsies and autopsies: fixation, embedding, staining Cytology: fixation, staining Immunohistochemistry and immunocytochemistry or Flow cytometry Molecular pathology: Generally, fluorescence in situ hybridization (FISH), chromogenic in situ hybridization (CISH), and polymerase chain

reaction (PCR) o Cytogenetics Image Management. Gross, histology and cytology slides (including digital slides), and molecular pathology images. This part of the work is performed jointly by the technicians, biologists and pathologists. Managing requests for special techniques, immunohistochemistry and molecular studies.

• • • •

Microscopic study. Interpretation and diagnosis. It is performed by the pathologist. In cytology, it usually enlists the help of cytotechnologists. Coding, which summarizes several specimen related codes (topography), procedures, findings or observations, and final diagnoses. Planning and work distribution. Management of information (reports, statistics, lists).

M. García Rojo /State ofthe Art and Trendsfor Digital Pathology

19

•

Decision-making assistance, clinicopathological correlation and cytohistological correlation. Assisted diagnosis, prognostic assessment, and therapeutic optimization.

• • • •

Pre-treatment of tumor registry, biobank and tissue bank data. Quality control of technical processes and diagnostics.

All these ñinctions can be grouped into: Generating information and integrating data from other sources. The integration of clinical, radiological and pathological data, requires a wellstructured collection of information and two-way communication, in order to keep the data updated. • Image Management. In pathology, imaging sources (modalities) are múltiple and heterogeneous, for capturing images both during specimen processing and during image interpretation. • Control of technical processes (quality of staining, amount of obtained DNA) and information quality assessment (correctly identifying containers, tissue blocks and slides). The collection of this information is performed through traceability systems. The results of participation in extemal quality assurance programs should be considered. The exploitation of these data allows continuous processes improvement. From all these ñinctionalities, indicators are obtained, statistics can be performed, and data mining is feasible, so that all this information allows for adequate resource planning. To do this, we need a standardization of indicators to be assessed (uniform specimen catalogue or complexity associated with each type of study). In Pathology, the complexity of a study is directly correlated with final pathologic diagnosis, so it is essential that there is a consensus in pathology diagnosis coding, establishing a pathology subset, based on SNOMED CT (Systematized Nomenclature of MedicineClinical Terms). The final product generated by the pathology department is the pathology report (biopsy, cytology, autopsy or molecular pathology). This report, once it is validated, it becomes part of the patient's medical record, as well as the associated images (or links thereto) that may be included within the report. Standardization in the pathology report structure should and how they should be sent to the electronic medical record or to public health records, such as population-based cáncer registries, are aspects that are already being addressed by Integrating Healthcare Enterprise [11]. IHE has published a technical framework that defines what structure should have

the standardized reports of pathology, based on HL7 clinical document architecture (CDA) standard [12]. 2.1. External Interoperability

The anatomic pathology information system (APIS) must be integrated with other information systems used outside the pathology department: • Electronic medical records in primary care and hospitals (see section 1. Electronic pathology study orders). • Teraiinology Servers. Usually based on SNOMED CT, they maintain controlled vocabularies of specimens, surgical procedures, service portfolio, observations and clinical and pathological diagnoses up to date.

20

M. García Rojo/ Stateof theArt cmd Trends for Digital Pathology

•

Image servers. Based on a picture archiving and communication system (PACS). Both images like still images (macroscopic or microscopic) and the digital slides should be available on a central server.

All the healthcare workflow related to pathology (surgery planning, intraoperative frozen studies planning, requests for cytology or autopsy studies, the selection of clinical data particularly relevant to the study, etc...) must be integrated with the hospital clinical information system (electronic medical record) and the pathology information system.

2.2. Internal Interoperability

The APIS must also be integrated with other information systems, usually existing in the pathology department:

• •

•

Automatic laboratory devices (devices of conventional and special stains, immunostainers). Work orders are generated and controlled from the APIS. Digital audio file recording system or speech recognition. It is used for word transcriptionof descriptions in gross and microscopic sections of the reports. Telepathology (telemedicine) portal. It can be used for primary remote diagnosis, teleconsultation, second opinion, and discussion forums. This is a distributed platform that can also collect reports from one or more APIS (in case of regional portáis).

•

•

Hospital-based tumor registry. Each tumor is registered with the information initially provided by the APIS, which then are supplemented with other essential information provided by the EMR or from other departmental systems such as Oncology or Radiotherapy Department. Biobanks (tumour bank). In those specimens collected for research purposed, it is essential to collect essential clinical and pathology data. Integration with the APIS improves the traceability of all specimens that have been used for research.

•

Quality Management. Quality assurance software maintains records of

incidents and provides documents, protocols, and standard operating procedures, which should be accessible also from the APIS.

•

• •

Bibliographic databases. Hyperlinks and efficient searches in Medline type databases and consensus clinical guidelines, like Bethesda 2001 for cervical cáncer, Bethesda 2007 for cytopathology of thyroid, College of American Pathologists cáncer protocols, and the consensus guidelines of the national societies of pathology). Picture libraries and repositories of cases of interest for educational purposes. Data warehouse and data mining. In addition to providing updated information on timing and statistics, this integration can improve delivery times and quality of diagnostics, by providing clinicopathological correlations (comparing initial and final clinical diagnoses with the pathology diagnosis) or cytohistological correladon (analyzing diagnoses from cytology reports and from surgical specimens reports). This analysis can also outline trends to improve forecasting of human and materials resources.

M. García Rojo /State of the Art and Trendsfor Digital Pathology

21

3. Traceability in the Management of Specimens and Intermedíate Products

3.1. Traceability during the Electronic Pathology Study Order

The request for a pathology stxidy requires the correct identification of the specimen whose study is desired, associating it with the right patient. During this step, in many cases, no specimen-identification label is printed (they will be printed when they arrive at the Pathology Department). Therefore, to maintain traceability a practical solution in these cases is printing patient-identification labels with bar code (preferably two dimensional, with the medical record number) from the request manager. The operating room management module, integrated with the request manager, must keep record of how many specimens have been generated to be sent to pathology and any incidence that may occur during the specimen collection or transportation. 3.2. Receiving the Specimen Each container sent to pathology is associated with an online request. Theoretically, a well labeled container can travel from the operating room or from the outpatient clinic to the pathology department, without being accompanied by any paper form. In practice, in many centers a pathology study request form is printed, in order to perform patient identification and specimen identification checking using the information from the printed form, instead of using computer screens to validate received specimens, or because the printed form of the pathology study request is used by the pathology department itself to make annotations during the macroscopic study or about tissue cassettes identification or list of techniques to be performed. Specimen reception is managed from the information system pathology. This is a critical step because in this moment the adequacy of the integration between clinical request manager and the APIS is tested. The verification is performed by the pathology technician and is essential to verify that all containers have arrived with appropriate identification and conservation. If not, all incidences should be recorded.

3.3. Registration Biopsies, Cytology, Molecular Pathology and Autopsies Once the specimen has been accepted by the pathology department, the next step is recording the study and assigning an internal (pathology) accession number that uniquely identifies the received specimen, usually through a sequential number (e.g. 11B0011123) and sometimes a letter, if the request study includes several containers (e.g. 11B0011123-A for breast quadrantectomy and 11B0011123-B for the lymphadenectomy part of the same surgical procedure). The APIS should be able to generate labels for containers, including patient's ñame, the date of receipt, the specimen type or organ, and the unique identification number of the container both in a readable format and barcode (dimensional or two-dimensional). 3.4. Specimen Gross Study

In this phase, cassettes labels are printed. These cassettes will become later into the paraffin blocks, which are the basic unit of tissue that usually manages a pathology department. The block (cassette) printing is done using the APIS or traceability

22

M. García Rojo /State of the Art and Trendsfor Digital Pathology

software, by the pathology technician. The label may include a two-dimensional bar code, plus the block number clearly legible. Since tissue blocks can be sent to other centers, it is interesting to also include the institution ñame or abbreviation on the label of the block, e.g. HCRl IBOOl 1123-A2.

3.5. Making Tissue Blocks, Microtomy, Slide Staining and Mounting

Several traceability systems for pathology laboratory are available in the market. They include three essential components for the paraffm block processing room and microtomy are: touch screens, bar code readers and slide label printers. The slide label printing can be performed directly on the slide or by special adhesive label printing that resist chemicals and physical processes to be performed on the histological and cytological slides. It is advisable to intégrate the tracking system as a module of function of the APIS.

3.6. Automatic Laboratory Devices

The number of automatic devices to be found in pathology services is increasingly growing and it usually includes tissue processors, automatic conventional techniques strainers, immunostainers, coverslips mounting, low-density hybridization array readers, real-time PCR, flow cytometers, extraction and nucleic acid amplification, DNA sequencing, pyrosequencers, one step nucleic acid amplification (OSNA) for sentinel node.

However, to date there is no standard for communication between these automatic

devices and the APIS. In many cases, ad hoc HL7 messages have been created after the mutual agreement between manufacturers. In the near future, IHE will also address the standardization of these messages. Major automatic staining manufacturers have systems to manage conventional staining (haematoxylin-eosin), special stains (e.g. PAS) and immunohistochemistry requests directly from the APIS. 3.7. Microscopic Description and Diagnosis

If microscopic examination is performed by conventional microscope, the pathologist receives a tray with cytological or histological slides in their office. It is recommended that the pathologist use the barcode reader (a bar code is available in the slide label) to edit the study reports, to avoid errors and to ensure that the text description or diagnosis issued after microscopic examination, is always associated with the appropriate report and patient. The request for additional techniques can also be performed using the slide barcode.

If microscopic examination is performed using digital slides, the pathologist enters the list of daily work and he checks for the reports that are pending. As he enters in each report, there is a direct access to digital slides, which appears on the same screen or on an accessory high resolution screen, coupled to the same computer.

M. García Rojo / State of the Art and Trends for Digital Pathology

23

4. Digital Image Management in Pathology

Some medical specialties (dermatology and hematology, etc.) require direct access to the macroscopic and microscopic images generated in pathology, in order to properly assess a clinical problem. Therefore, pathology images should also be part of the patient's health record. The efficient management of pathology images should follow a similar model, but not identical, to that used in radiology and must begin with a request to create the image in a specific device or method, to generate a working list of all the requests, and must end with the notification of acknowledgement of storage in the PACS. All this must be managed centrally from the pathology information system (APIS). However, since the image acquisition devices in pathology are not compatible with the medical imaging standard DICOM (Digital Imaging and Communications in Medicine), it requires an additional step of image "dicomization", so that they can be stored in the PACS.

Unlike radiology images are specimen centered and not patient centered, so the image performing requests origínate only within the pathology department, through the APIS.

The proposed scheme in image acquisition is as follows: • The APIS creates the work list of images to be generated. • A middleware (on the telepathology website) sends this Hst to the corresponding modality or device. • The pathologist or technician checks the list and generates the image (macroscopic or microscopic), which will already have associated identification data.

• •

After confirming the quality of image it is sent to PACS. The telepathology portal receives confirmation of proper storage in the PACS and this information is transmitted to APIS.

This scheme works well when the images are small (JPEG-based macroscopic systems or microscopic photos). In the case of digital slides, where a single slide can have a size of more than 1 GB, even after efficient compression, a specific solution must be available for these very large files. DICOM has two altematives. The first is to use special servers for large images (type JPEG2000) [13]. The second solution proposed by the working group 26 of DICOM is to fragment the large images in thousands of small fragments, which then can be stored on the same servers that are already storing radiological images [14]. However, with such a high number of fragments of images that need to be managed by the PACS, it is possible that the PACS performance can become compromised. This is an important decisión, because if a special server for pathology images is installed, in order to improve performance, maintenance costs of servers in the computing department can become significantly increased. Today, the main limitation of digital imaging in pathology is the scanning speed of digital slides. Existing scanners will usually require an average of about 15 minutes to sean a single slide at 40x magnification if it is not a small specimen. Therefore, in general, you can sean only about 100 slides daily with a single scanner. Given that médium size pathology departments generates about 500 slides per day, including cytology, biopsy and autopsy, about 5 slide scanners would be needed to assume daily workload. Fortunately, the scanning technology is evolving very quickly and there are

24

M. García Rojo/ State ofthe Art and Trendsfor Digital Pathology

some systems on the market that are able to sean a single preparation with 40x magniñcation in about one minute.

The storage requirements for the management of digital pathology images can become an important limiting factor, but technologically very efñcient solutions aheady exists. We have caleulated that about 100 TB of storage per year would be needed to store all digital slides in a pathology department in a 500 beds hospital. As for the display of digital slides, since the current DICOM viewers, used to display radiological images, do not have the tools to navigate through large images, specific viewers of pathology digital slides must be used. We recommend using light web-based viewers. This will avoid having to install slide viewer software in every

computer where they these images need to be displayed. This allows easier access to digital slides for specialists such as dermatologists and haematologists. Microscopic digital images allow thepathologist to use tools that can be useful for image interpretation. FDA-approved algorithms for automated analysis of digital pathology images, to automatically quantify the immunohistochemical expression of some biomarkers such as oestrogen receptor, progesterone receptor, Ki67 proliferation Índex, Her2-Neu, are already available. All these markers are fundamental today, for instance, in the diagnosis and treatment breast cáncer. Other developing algorithms allowdetection of suspicious areas of cáncerin a histological or cytology slide [15]. In

some applications, such as the study of Parkinson's disease, Alzheimer's, or early prostate cáncer, the possible role for three-dimensional reconstruction of microscopic images from digital slides is being assessed. To address these research studies today, massively parallel processing architectures based on grid services are being used [16]. Figure 1 shows the architecture of image management and pathology report managing, integrated with electronic medical records and telepathology portal, based on the model used in the Health Service of Castilla-La Mancha (SESCAM) in Spain.

Tedinidan p

P I

Digital Slide Storage

Gross station Macroscopy stHIímages

Miaoscopy si Oiagnosi

Figure 1. Integral management of images andreports in anatomic pathology

M. García Rojo / State of the Art and Trendsfor Digital Pathology

25

5. The Telepathology Portal

The result of pathological activity (reports, images and second opinion) from of all institutions in the same health service can be centralized, using a telepathology portal, which collects all data and images from pathology departments. Given the patient geographic mobility, a single query point is needed where all the pathology reports and associated images are available. This is especially advantageous for clinicians and can also be helpful for the implementation of summary-level medical record nationally and intemationally. In addition, this platform is a meeting point for all pathologists in the same health service, making it suitable for use as teleconsultation, telepathology service for either primary diagnosis (e.g. intraoperative frozen studies) or for second opinion. Telepathology portal is possible even in those health services where hospitals have heterogeneous pathology information systems, because using the IHE recommendations, central repositories of pathology reports can be created, either by sending a copy of the pathology report (e.g.. sending a PDF document to the repository, that should be maintained up to date), or in a distributed query, that is launched to the different pathology databases of every hospital, which dynamically generates the requested pathology reports. Those reports submitted to the telepathology portal should be encoded with SNOMED CT, especially for the type of specimen and the fmal diagnosis. In the case of telepathology portal of the SESCAM (Spain), the portal contains the following areas [17]: • Scanning. To view images or slide scanning request management (for those centres that do not have a slide scanner). This includes macroscopic and microscopic still images and digital slides. • Reports. Inquines for pathology reports of all partner institutions, with flexible criteria (patient ñame or súmame, personal identification code, centre, record number of each centre, etc.).

•

•

Discussion. Allows second opinion and teleconsultation to other pathology departments in the heaUh service (usually directed at a particular consultant). A public forum is also available, where all pathologists can answer and particípate. Also, tumor board meetings and clinicopathological sessions can be arranged, intra-hospital or in a regional context. Training and knowledge management. Since all information and images that the site is managing is perfectly structured and encoded, it can tum into valuable teaching and research material, once the patient identification data is dissociated. This creates an interesting case library, atlases and online surveys for undergraduate teaching and pathology residents training.

6. Semantic Interoperability

In IHE, the defmition of a proposed workflow for pathology services, which was defmed in Volume 1 of the technical framework of Pathology [11], implies that pathologists should use a common nomenclature for such everyday objects as received specimens, containers, tissue fragment that is selected for histopathological study, or each of the fragments that appear in a histological slide.

26

M. García Rojo / State ofthe Art atid Trendsfor Digital Pathology

Semantic standardization in pathology has been extended to: • Identification of procedures to be performed; • Identification of specimens; • Observations included in the pathology report; • Diagnoses. In order to achieve these goals, three terminologies can be used: LOINC (Ñames and Codes Observation Identifiers Logic), SNOMED CT and PathLex [12]. It is necessary to train pathologists in SNOMED CT. They should abandon the classical encoding SNOMED II systems, using topographic (T), procedures (P) and

diagnostics (M) listings, since this system is no longer maintained by the Intemational Development Health Terminology standards (IHTSDO). Instead, SNOMED CT should be used, since this system is based on relationships between terms, and it contains more than 311.000 unique concepts. In those pathology services in which SNOMED II or a proprietary coding system is being used, those codes should be mapped and converted into SNOMED CT codes, so that all information can be exchanged in a structured manner (for example, to send pathology diagnosis the EMR) using a universal concept ID of SNOMED CT for each term.

7. Conclusions

Information systems in pathology will evolve significantly along with advances in molecular biology and in microscopic image scanning systems. The digitization of histological and cytological slides in pathology allows that such images are also accessible from the patient's medical record. This solution will make easier for other specialists, like dermatologist or haematologist, to access these images, and it will have a favorable impact in medical students and residents training. Using teleconsultation and remote diagnosis based on digital slides may alleviate, at least in part, the shortage of speciaHsts in some countries. These telepathology services will grow rapidly when all pathology images are available in a standard digital format.

In

research, this will mean a revolution

in the

validation of new

pathophysiological concepts and therapies. Nowadays, digital slides are a major challenge to existing health informatics

standards. Implementing the DICOM image standard, also available now for digital slides, is necessary for long term interoperability solutions. Standardization efforts are bearing fruit also in digital pathology. This is the way for the interconnection between different telepathology networks to access, exchange and update data in the patient medical record, through the Internet. Digital pathology today offers an unprecedented flexibility in the delivery of pathology reports, since it is possible to generate different types and report formats for different users with different levels of understanding.

The digital pathology solutions integrated with electronic medical record and centralized images storage will improve the cooperative work between clinicians and pathologists, in order to increase patient safety and quality of health services. Perhaps in the future is will also be necessary to obtain three-dimensional information of the histological and cytological slides. 3-D navigation, the performance improvement of image viewers, new tools to identify areas of interest, and even the

M. García Rojo / State ofthe Art and Trends for Digital Pathology

27

appearance of specific hardware, will make pathologists to replace the microscope with a new device that will allow a more efficient and better quality work.

References

[1]

[2]

[3]

R. Ferreira, B. Moon, J. Humphries, A. Sussman, J, Saltz, R. Miller, A. Demarzo, The Virtual Microscope, Proceedings of the AMA Anmial Fall Symposium (1997), 449-453. Available at: http://www.piibmedcentral.nih.gov/articlerender.fcgi?tool=pubmecí&pubmedid=9357666 A. Afework, M.D. Beynon, F. Bustamante, S. Cho, A. Demarzo, R. Ferreira, R. Miller, M. Silbemian, J. Saltz, A. Sussman, H. Tsang, Digital dynamic telepathology - The Virtual Microscope, Proceedings of the AMIA Symposium (1998), 912-916. Available tat: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2232135 J. Zeineh, Reinvention of light microscopy. The Trestle Digital Backbone, In: J. Gu, R.W. Ogilvie, R.W. (eds.), Virtual microscopy and virtual slides in teaching, diagnosis, and research. Series Advances in Pathology, Microscopy & Molecular Morphology, Taylor & Francis, Boca Ratón, FL, USA, 2005, 77-96.

[4]

[5] [6] [7]

[8]

[9] [10]

[11]

[12]

[13]

[14]

[15]

[16] [17]

J.R. Gilbertson, J. Ho, L. Anthony, D.M. Jukic, Y. Yagi, A.V. Parwani, Primary histologic diagnosis using automated whole slide imaging: a validation study. BMC Clinical Pathology, 6 (2006). Available at: http://www.biomedcentral.eom/1472-6890/6/4 F.R. Dee, A. Donnelly, S. Radio, T. Leaven, M.S. Zaleski, C. Kreiter, C, Utility of 2-D and 3-D virtual microscopy in cervical cytology education and testing. Acta Cytologica 51 (2007), 523-529. D.P. O'Malley, Practical applications of telepathology using morphology-based anatomic pathology. Archives ofPathology and Laboratojy Medicine 132 (2008), 743-744. K. Kayser, J. Szymas, R.S. Weinstein (eds.), Telepathology. Telecommunication, electronic education and publication in Pathology, Springer, Berlin, 1999. T. Sawai, The state of telepathology in Japan, In: T. Sawai (Ed.), Telepathology in Japan. Development andpractice. Cele, Inc, Morioka, Twate, Japan, 2007, 3-9. D.F. Cowan, (ed.), Informatics for the Clinical Laboratory. A practical guide for the pathologist, Springer, New York, 2005. College of American Pathologists, Anatomic pathology computer systems, CAP Today. February (2011), [on line]. Available at: http://captodavonline.com/productguides/software-svstems/anatomicpathologv-computer-svstems-cap-todav-februarv-2011 .html IHE Intemational, Anatomic Pathology Technical Framework. Volume 1 (PAT TF-1). Profdes. Revisión 2.0, [on line], 2010. Available at: http://www.ihe.net/Technical Framework/upload/IHE PAT TF Rev2-O Voll_TI_2010-07-23.pdf IHE Intemational, Anatomic Pathology Structured Reports (APSR), [on line], 2011. Available at: http://www.ihe.net/Technical_Framework/upload/IHE PAT Suppl APSR Revl1 TI 2011 03 31.pdf V.J, Tuominen, J. Isola, Linking whole-slide microscope images with DICOM by using JPEG2000 Interactive Protocol, Journal of Digital Imaging, 23 (2010), 454-462. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2896636 DICOM Standards Committee, Working Group 26, Pathology, Supplement 145: Whole Slide Microscopic Image lOD and SOP Classes, [on line], Virginia, USA, 2010, Available at: ftp://medical.nema.org/medical/dicom/fínal/sup 145 ft.pdf G. Bueno, R. González, O. Déniz, J. González, M. García-Rojo, Colour model analysis for microscopic image processing, Diagnostic Pathology, 15 Suppl 1 (2008), SI 8. Available at: http://www.ncbi.nlm.nih.gOv/pmc/articles/PMC2500098 G. Bueno, O. Déniz, J. Salido, M.G. Rojo, Image processing methods and architectures in diagnostic pathology. Folia Histochemica et Cytobiologica, 47 (2009), 691-697. M. García Rojo, C. Daniel, Digital Pathology and Virtual Microscopy Integration in E-Health Records", in: S. Mohammed, J. Fiaidhi (eds), Ubiquitous Health and Medical Informatics: The Ubiquity 2.0 Trendand Beyond, IGI Global, PA, USA, 2010, 457-484.

' IHTSDO®, SNOMED® and SNOMED CT® are registered trademarks of the Intemational Health Terminology Standards Development Organization.

M. García Rojo / State of the Art atid Trendsfor Digital Pathology

DICOM® is the registered trademark of the National Electrical Manufacturers Association for its standards publications relating to digital communications of medical information. HL7® is a registered trademark of Health Level Seven (HL7), Inc. IHE® is a registered trademark of Integrating the Healthcare Enterprise (IHE) Intemational. LOINC® is a registered trademark of Regenstrief Institute, Inc. UMLS® and Metathesaurus® are registered trademarks of the National Library of Medicine.

Perspectives on Digital Pathology

29

M. García-Rojo et ai (Eds.) IOS Press, 2012

© 2012 The authors and IOS Press. All rights reserved. doi: 10.3233/978-1 -61499-086-4-29

State of the Art in Telemedicine - Concepts,

Management, Monitoring and Evaluation of the Telemedicine Programme in Alentej o (Portugal) Tiago Cravo OLIVEIRA"' Maria José BRANQUINHO" and Luís GONCALVES"' ^Imperial College Business School, Imperial College London. United Kingdom

^Administragao Regional deSaúde do Alentejo, LP. Portugal Abstract. Alentejo - one of five Portuguese continental regions - faces major problems impacting the health and social system of the región. Here, the low population density, the low educational and income level as well as an aging population have to be mentioned. Faced with the task of ensuring equal access to healthcare for all its inhabitants, the regional health authorities created the telemedicine program.

From 1998 until 2000, the program developed in an experimental fashion, with teleconsultations involving a number of providers; primary health care centers, regional hospitals, and central hospitals. Between 2000 and 2010, there were a total of 135,000 telemedicine acts including teleconsultations, teleradiology (computerised tomography and x-rays), ultrasound telemedicine and telepathology. Presently, the network comprises 20 health centers and 6 hospitals, covering 4 districts. The platfonn is composed of high resolution videoconferencing equipment, software with patients' clinical records, an image archive, and a number of peripherals, such as electronic dermatoscopes and phonendoscopes. Teleconsultations are provided by fifteen medical specialties, across 3 district hospitals, ranging from neurology to pediatric surgery. In 2008, health authorities started the teleleaming program, initially using point to point videoconferencing, and by the end of 2010, 848 healthcare professionals, across 52 locations, liad participated in remote leaming sessions, covering topics from chronic wound treatment, to infection control, to medical error. As of 2011, point to multipoint teleleaming is also in operation. This paper provides an overview of the telemedicine program in Alentejo, including both infrastructure and operations. Preliminary results of an ongoing

evaluation of the impact of teleconsultations on key indicators of the regional healthcare system are also presented (including current utilization and plans for future expansión). This article builds on the experience acquired throughout a decade of using telemedicine on an everyday basis, in a context of remarkable challenges in the delivery of accessible, equitable and quality healthcare services. Keywords. Telemedicine, teledermatology, telepathology, telehomecare, teleleaming, concepts, management, monitoring, evaluation, Alentejo.

Corresponding Author. Luis Gon^alves [email protected]

Administra9ao Regional de Saúde do Alentejo. Portugal.

30

T.C. Oliveira et al. /State of the Art in Telemedicine

Introduction

Alentejo - a región of continental Portugal - represents one third of the country's

territory, i.e. 23,742 km^. This región accommodates only 5 per cent of Portugal's population, resulting in an average of 19.53 people per square kilometer. A quarter of Alentejo's inhabitants is over 65 years oíd and only 5 per cent have a higher education degree. In 2007, the region's per capita income was 73 per cent of the European Uniones average. An ageing population and low leveis of education and income create considerable social and economic challenges for local policy makers and regional authorities. This is especially true in the provision of equitable and accessible healthcare services to the region's population. Based on the fundamental idea that patients should be seen, diagnosed and treated as cióse as possible to where they live and work, the regional health authorities started using telemedicine as a tool to achieve that proximity.

The expectations were that the telemedicine program would lead to a reduction in the distances between healthcare services and patients, a reduction in unnecessary travelling between specialized units and patients, a shortening of waiting times, and fmally support for healthcare workers and population living in remote, underserved locations [1-4]. The programme's objectives were to: • Increase the accessibility of specialist outpatient appointments; • Ensure equity in the access to the best available care for all patients; • Reduce the costs of publicly funded and provided healthcare services; • Reduce the distance between primary and secondary care.

1. The Telemedicine Programme

From 1998 to 2000, the telemedicine program developed in an experimental fashion. The first teleconsultations involved health centers, regional and central hospitals. Teleconsultations in Alentejo aimed at connecting the general practitioner and the patient, who are based in the local primary healthcare centre, to the hospital consultant via videoconferencing equipment. Following initial encouraging results, the program took off, and between 2000 and 2010 a total of 135,000 telemedicine acts were

performed, including real-time teleconsultations, teleradiology (computerized tomography and x-rays), ultrasound telemedicine and telepathology. Since the start of the program, the number of acts has increased steadily, as demonstrated in Figure 1. For face-to-face outpatient appointments, teleconsultations have to be requested by the general practitioner. In the actual consultation, there are two physicians instead of just one (the general practitioner and the hospital consultant). Patient data is readily available using an electronic patient record. Like in a traditional face-to-face consultation, the consultant can request diagnostic procedures. Once a diagnosis is made, treatment can be prescribed usually by the general practitioner, or in certain cases by the consultant (e.g. certain medications have to be prescribed by a neurologist). If it is not possible to make a diagnosis remotely, then a face-to-face consultation is scheduled. The number of patients who require a subsequent face-to-face appointment following a teleconsultation varies between different specialties. Initially, requests for teleconsultations were communicated using fax [1]. This is now changing to an electronic booking platform. At present, there are three hospitals electronically

T.C. Oliveira et al. /State ofthe Art in Telemedicine

31

scheduling teleconsultations (Portalegre, Évora and Elvas). The schedules are managed by both the consultants and administrativo staff, which have also received training in

operating the videoconferencing equipment. This way, the consultant is freed to focus on his main task: taking care of the patient.

S

g

-

(# c#

c?»'" ^

é" ^

Figure 1. Telemedicine acts in Ale;ntejo from 1998 to 2010

Currently, there are 26 telemedicine platforms distributed between 20 health

centres and 5 hospitals, covering 4 districts in Alentejo: Portalegre, Évora, Beja and Litoral Alentejano. In the early years of the programme, teleconsultations between regional and central hospitals (in genetics and cardiology) have been performed, which those are now at a standstill. However, teleradiology is still performed between regional and central hospitals. Teleconsultations between the four regional hospitals and the health centres include a variety of medical specialties and purposes: diabetes, traumatology, orthopedics, general and pediatric surgery, pneumology, urology, gastroenterology, clinical oncology, cardiology, dermatology, neurology, physical medicine and rehabilitation, pain, thyroid and obesity. The largest number of teleconsultations is performed by dermatology, neurology and gastroenterology. There are also recent efforts in real-time teleconsultations in anatomical pathology (the región is involved in the development of a European telepathology network) [2,4].

2. The Technological Infrastructure

The telemedicine platform infrastructure is based on the Health Information Network (Portuguese: Rede Informática de Saúde), a high speed virtual private network

32

T.C. Oliveim et al. / State of the Art in Telemedicine

managed by the Ministry for Health. Despite the high speed communication available in hospitals, health centers frequently manage only a very limited bandwidth (especially more remote units), which greatly limits the quality and rehability of telemedicine services. The network is provided free of charge to healthcare units in the Portuguese National Health Service (NHS). The telemedicine platforms include high resolution videoconferencing equipment,

software which allows access to the electronic patient record and picture archive, and peripherals such as electronic phonendoscopes and dematoscopes. For 2012, there are plans to increase the number of peripherals in the health centers and install ultrasound equipment. Over a period of ten years, the telemedicine program has seen investments totaling €968,000. This is expected to increase in the near future, with plans for telehomecare and expansión into new primary care health centers [1-4].

3. The Telelearning Program

The telelearning program started in 2008, using point to point videoconferencing [3]. The equipment is the same as that for teleconsultations, which promotes a more efñcient use of resources. In 2011, with the acquisition of new equipment, it became possible to perform point to multipoint ti'aining sessions. The telelearning program seeks to:

• Minimize the impact of the region's geography on training; • Reduce the costs associated with traditional training programmes; • Increase the motivation of healthcare professionals; • Provide health education to populations. Since 2009, annual plans for telelearning sessions have been developed. The sessions are publicized in posters distributed through the region's health centers, as well as by e-mails sent to those who are expected to benefit from specific sessions. In order to identify these individuáis, a database was set up which not only facilitates the planning of fiiture sessions, but is also used to monitor and register attendance at each session. Figure 2 illustrates the posters fi"om 2009 to 2011.

PROGRAMA

DcTELEMEDfCINA

DO ALENTEJÓ

AMA

rr TELEFORMACÁO ALENTEJO

p-.v

006)^ TRArfinMTHI^

Figure 2. Posters for the 2009, 2010 and 2011 telelearning annual programs

T.C. Oliveira et al. / State of the Art in Telemedicine

33

Tables 1 to 3 show the attendances and number of locations for each of the

teleleaming sessions between 2009 and 2011. Presently, a total of 848 professionals have participated in teleleaming sessions (this exeludes the last session of 2011). Table 1. Teleleaming sessions in 2009 - attendance and number of locations

Attendance

Session

Number of locations

Informed consent

72

2

Clinical risk

63

2

Telephone calis

105

7

Infection control

91

7

Wounds

71

7

Total number of professionals attending: 402

Table 2. Teleleaming sessions in 2010 -- attendance and number of locations Attendance

Number of locations

Bronchial asthma

13

2

Clinical risk

38

2

Clinical examination of children

11

2

Clinical examination of the new-bom

12

2

Infomied consent

47

2

Session

Total number of professionals attending; 121

Table 3. Teleleaming sessions in 2011 - attendance and number of locations Attendance

Number of locations

Infection control: hygiene in healthcare units

42

4

Infection control: cleaning and disinfecting

44

6 (3 simultaneous)

Bronchial asthma

30

5 simultaneous

National accreditation programme

59

8 simultaneous

Endocrinology congress

68

8 simultaneous

Informed consent

110

8 simultaneous

Session

Total number of professionals attending: 325

34

T.C. Oliveira et al. /State ofthe Art in Telemedicine

Telelearning sessions usually include an initial exposition from the trainer and subsequent discussion of real cases. Every year, participants' satisfaction is surveyed. Figure 3 illustrates the high levels of overall satisfaction with telelearning sessions (with satisfaction improving from 2009 to 2010). Lower levels of satisfaction related to scheduling and documentation provided during the sessions. Even so, in 2010, every participant surveyed stated that he/she would advise colleagues to particípate in telelearning sessions in the future.

3%

Overall satisaction In 2009

• Veryiow • Low •Médium

3%

Overall satisaction in 2010

•High

iVeryhigh

Figure 3. Overall satisfaction with the telelearning sessions in 2009 and 2010

4. Healthcare Professionals' Views of Teleconsultations

The impact of the teleconsultations program is presently being assessed in a comprehensive manner. The results from this evaluation are to be published before the end of next year. This section reports about a qualitative study on healthcare professionals views of teleconsultations, conducted in November 2010 [5], The study was based on semi-structured interviews with various physicians and primary care managers in Alentejo who were, or had been, involved in the telemedicine program. The participants included seven hospital consultants (dematology, cardiology, psychiatry, physical and rehabilitation medicine, neurology and gastroenterology, general surgery), two primary care managers, and three general practitioners. Participants were first asked about the main difficulties in providing healthcare services in Alentejo. Typical answers included long distances between patients and healthcare units, limited public transportation, a shortage of hospital consultants given the size of the population, long waiting times and large numbers of consultations expected from hospital consultants. Secondary care services in Alentejo are indeed limited. For reasons that go beyond the scope of this paper, the región has a shortage of hospital consultants in a number of medical specialties (e.g. dermatology and neurology). Capacity is insufficient and hospital consultants ultimately see more patients than their counterparts in Lisbon or Porto. Long waiting lists effectively create a barrier to access, which is made worse by long distances and a poor public

T.C. Oliveira et al. /State ofthe Art in Telemedicine

35

transportation network. Remoteness affects not only patients, but general practitioners as well, who feel they are left out of the system, losing important opportunities to interact with other healthcare professionals. We were told that teleconsultations are effectively a tool which minimizes some of

these difficulties. The advantages are clear, according to the physicians that use the technology. The average teleconsultation is shorter than the average face-to-face consultation, because small talk is reduced to a minimum and the general practitioner's

presence streamlines the process. This is in accordance with existing evidence and results in a bigger service capacity and productivity gains [6-9]. The benefits for patients and families are irrefutable and in line with previous reports: shorter waiting times [7, 10, 11], shorter distances [10, 12-15], fewer costs from travelling and time off work [9, 10, 13, 14, 16-21]. There is also the potential for increasing the quality of referrals. Since general practitioners are present during teleconsultations, they effectively receive training on the job [9, 11, 22-28]. In other words, every teleconsultation has the potential to involve some level of teleleaming. This not only affects general practitioners' ability to deal with more patients in primary care and potentially reduce referrals, but it also improves the overall quality of those patients that do get referred. A benefit for patients is the avoidance of travelling and costs associated with unnecessary or inappropriate referrals. A further benefit for general practitioners is the increased information on what happens during the specialist's appointment. While teleconsultations have the potential to reduce referrals, there is also the

potential for duplication of services. We were told that, depending on the specialty, as much as a fifth of teleconsultations resulted in a subsequent face-to-face appointment. Reasons for this included: suspicion of severe condition, complications and resistance to therapy, need for face-to-face procedures, and technical difficulties with the equipment. There is valué in having a teleconsultation before a face-to-face consultation, since the hospital consultant has already seen the patient and knows what to do when he arrives for the face-to-face appointment. However, it is arguable whether this applies to all cases, and whether some patients should be sent directly to the hospital consultant rather than having an initial teleconsultation. On the issue of utilization, there was another interesting fmding. Several general practitioners considered that patients who are longer waiting for a hospital appointment tend to visit the primary healthcare centre more often. Teleconsultations may reduce this potentially unnecessary use of services, due to shorter waiting times. Asked about how teleconsultations affected their prescribing, most physicians told US that there were no differences. The inability to physically touch the patient did not seem to affect prescribing either way. On a different topic, the majority of participants believed that teleconsultations could substitute for as much as 60 percent of all face-toface consultations in the región, and that this would be highly beneficial. The actual valúes varied according to the specialty: for dermatology, for example, we were told that for health centers more than 30 kilometers away from the hospital, all patients could be seen through teleconsultations. All participants said they would advise colleagues to use teleconsultations, with a few caveats mentioned. If it is possible to have a face-to-face consultation (i.e. if the distance to the service is small) then there is no need for a teleconsultation. Furthermore, one participant considered that teleconsultations should be performed by experienced consultants.

36

T.C. Oliveira et al. /State ofthe Art in Telemedicine

5. Future Work

Work is underway to improve the bandwidth and reliability of the Health Information Network. A plan to expand the infrastructure has also been approved. This will increase the number of platforms available throughout Alentejo as well as the number and quality of peripherals. The teleleaming program will continué in 2012 with an emphasis on improving the fit between the needs of healthcare professionals and the content of the sessions.

An ISO 9001 certification of the telemedicine program is also underway and should be complete by mid-2012. By the end of next year, the teleleaming program is expected to be certified as well. The evaluation of the teleconsultations program, mentioned in the previous section, is also expected to produce results by mid-2012. Finally, there are plans to start a telehomecare program in the región. With an estimated 43 percent of the population over 65 years oíd in 2050 and many suffering from múltiple chronic conditions, telehomecare is expected to have considerable benefits.

6. Conclusions

The experience with the telemedicine program for Alentejo is extremely positive. There is little doubt that the program promotes equity and increases access to hospital outpatient care. The virtual elimination of distance as a barrier to access health services not only benefits the patient, but also general practitioners who are isolated and shut out of the system. The economic benefits for patients are also undeniable. The evaluation of the teleconsultation program currently under way is also expected to shed light into the economic benefits and costs for the NHS and healthcare units (which take the brunt of the costs and with more limited benefits). As for the teleleaming program, the economic benefits are clearer, from all perspectives. Compared to physical face-to-face training sessions, teleleaming saves considerable amounts of resources (both in time and money). Participant satisfaction is high and rising. With building up experiences from previous years, it is expected that further improvements in attendance and quality of the sessions will be achieved. Alentejo faces considerable challenges. The region's population is on average poorer, older and geographically more disperse than the rest of the country. It suffers further from an inability to attract a young active population, making it even more difficult to fmance the services an aging population needs. Under these conditions, telemedicine is an essential tool to address the healthcare needs of the population, improving equity and access at potentially lower cost. After an initial experimental period, the use of telemedicine is now an integral part of everyday's healthcare provision in Alentejo. The program is maturing, with accreditation and evaluation processes underway. And with plans for new projects (telehomecare), there is no doubt that telemedicine will play a fundamental role in shaping the future of the region's healthcare services.

References

[1]

Programa de Telemedicina do Alentejo. Unpublished report, L. Gongalves, Administra^ao Regional de Saúde do Alentejo, I.P., 2003.

T.C. Oliveira et al. /State of the Art in Telemedicine

37

[2]

Programa de Telemedicina do Alentejo. Unpublished report, L. Gonpalves, Administra9áo Regional de

[3]

Projecto de Teleformagáo no Programa de Telemedicina do Alentejo. Unpublished report, L.

[4]

SimulaDiagnóstica. Unpublished report, L. Gon9alves, Administra9áo Regional de Saúde do Alentejo,

[5] [6] [7]

Reportforfirst roundofdata collection and interviews. Unpublished report, T. Gravo Oliveira, 2010. A. García Rada, Virtual hospital forHIV patients halves consultation times, BMJ342 (2011), d1818. A. León, C. Cáceres, E. Fernández, P. Chausa, M. Martín, C. Codina, et al. A New Multidisciplinary Home Gare Telemedicine System to Monitor Stable Ghronic Human Immunodeficíency Virus-Infected

Saúde do Alentejo, I.P., 2008.

Gon9alves, M.J. Branquinho, Administra9áo Regionalde Saúdedo Alentejo, I.P., 2008. I.P.,2011.

Patients: A Randomized Study, PLoS ONE 6 (2011), el4515.

[8]

L. Baldwin, M. Glarke, L. Hands, M. Knott, R. Jones, The effect of telemedicine on consultation time, J Telemed Telecare 9 (2003), 71 -73.

[9]

R. Wootton, S.E. Bloomer, R. Gorbett, D.J. Eedy, N. Hicks, H.E. Lotery, et al, Multicentre randomised control trial comparing real time teledermatology with conventional outpatient dermatológica1 care: societal cost-benefit analysis, ^M/320 (2000), 1252-1256.

[10] K. Yogesan, S. Kumar, M.L. Tay-Keamey, I. Constable, Evaluation of Intemet-based eyecare delivery, J Telemed Telecare 12 (2006), 103-104.

[11] E.J. Nordal, D. Moseng, B. Kvammen, M.L. Lochen, A comparative study of teleconsultations versus face-to-face consultations, J Telemed Telecare 1 (2001), 257-265.

[12] P.T. Jaatinen, P. Aamio, J. Remes, J. Hannukainen, T. Koymari-Seilonen, Teleconsultation as a replacement for referral to an outpatient clinic,J Telemed Telecare 8 (2002), 102-106. [13] T.S. Bergmo, An economic analysis of teleradiology versus a visiting radiologist service, J Telemed Telecare 2 {\996\ 136-142.

[14] G. Sicotte, P. Lehoux, N. Van Doesburg, G. Cardinal, Y. Leblanc, A cost-effectiveness analysis of interactive paediatric telecardiology,J TelemedTelecare 10 (2004), 78-83.

[15] A.M.M Oakley, P. Kerr, M. Duffill, M. Rademaker, P. Fleischl, N. Bradford, Patient cost-benefits of realtime teledermatology-a comparison of data from Northem Ireland and New Zealand, J Telemed Telecare ^ {20m\ 97-101.

[16] M.A. Edwards, A.C. Patel, Telemedicine in the State of Maine: A Model for Growth Driven by Rural Needs, Telemedicine Journal and e-Health 9 (2003), 25-39.

[17] J. Stensland, S.M. Speedie, M. Ideker, J. House, T. Thompson, The Relative Cost of Outpatient Telemedicine Services, Telemedicine Journal 5 (1999), 245-256.

[18] Real-Time (Synchronous) Telehealth in Primaty Care.Systematic Review of Systematic Reviews [Technology report no 100], A. Deshpande S. Khoja A. McKibbon A.R. Jadad (eds), Canadian Agency for Drugs and Technologies in Health, Ottawa, 2008. Available at: http://cadth.ca/media/pd£^427A Real-Time-Svnchronous-Telehealth-Primarv-Care tr e.pdf [19] C.M. Cusack, E. Pan,J.M. Hook, A. Vincent, D.C. Kaelber, B. Middleton, Thevalué proposition in the widespread use of telehealth,J TelemedTelecare 14 (2008), 167-168.

[20] V. Wade, J. Kamon, A. Elshaug, J. Hiller, A systematic review of economic analyses of telehealth services using real time videocommunication, BMCHealthServices Research 10 (2010),233. [21] N. Eminovic, M. Dijkgraaf, R. Berghout, A. Prins, P. Bindels, N. De Keizer, A cost minimisation analysis in teledermatology: model-based approach, BMCHealthServices Research 10 (2010), 251. [22] J. Norum, S. Pedersen, J. Stormer, M. Rumpsfeld, A. Stormo, N. Jamissen, H. Sunde, T. Ingebrigtsen, M.L. Larsen, Prioritisation of telemedicine services for large scale implementation in Norway, J Telemed Telecare 13 (2007), 185-192.

[23] S. Jarvis-Selinger, E. Chan, R. Payne, K. Plohman, K. Ho, Clinical Telehealth Across the Disciplines: Lessons Leamed, Telemedicine and e-Health 14 (2008), 720-725.

[24] P. Taylor,Evaluating telemedicine systems and services, J Telemed Telecare 11 (2005), 167-177. [25] H. Lamminen, J. Lamminen, K. Ruohonen, H. Uusitalo, A cost study of teleconsultation for primarycare ophthalmology and dermatology. J Telemed Telecare 1 (2001), 167-173. [26] D. Lobley,The economics of telemedicine, J TelemedTelecare,(1997), 117-125. [27] W.P.M. Vierhout, J.A. Knottnerus, H.F.J.M. Crebolder, A.M.K. Wesselingh-Megens, G.H.M.I. Beusmans, A. Van Ooij, P. Pop, Effectiveness of joint consultation sessionsof generalpractitioners and orthopaedicsurgeons for locomotor-system disorders, TheLancet, 346 (1995), 990-994. [28] M.A. Loane, S.E. Bloomer, R. Gorbett, D.J. Eedy, C. Evans, N. Hicks, et al, A randomized controlled trial assessing the health economics of realtime teledermatology compared with conventional care: an urban versus rural perspective, J Telemed Telecare 1 (2001), 108-118.

38

Perspectives on DigitalPathology M. García-Rojo et al (Eds.) IOS Press, 2012

© 2012 The authors and IOS Press. All rights reserved. doi: 10.3233/978-1 -61499-086-4-38

Paradigm Changes in Health Lead to Paradigm Changes in Pathology Jacqueline A. HALL^'^ and Bernd BLOBEL^ ^European Organizationfor theResearch and Treatment of Cáncer, Brussels, Belgium eHealth Competence Center, University Hospital Regensburg, Regensburg, Germany

Abstract. For the sake of safety and quality of care as well as efficiency of care processes, health systems undergo a paradigm change towards personalized, ubiquitous, health services. This change includes preventive and predictive medicine based on advanced translational medicine. Here we introduce domain-

specific, organizational, and technical paradigms, requirements and solutions for personalized, ubiquitous, care. Emphasizing the formal aspects of modeling and implementing Telehealth and personal health (pHealth) interoperability and the entailed multidisciplinary integration, and illustrate the drivers behind and benefits of personalized medicine with a specific focus on the changingtrends and impact on pathology, especially emphasizing Telepathology.

Keywords. Personal Health, pHealth, telemedicine, systems medicine, digital pathology

Introduction

Facilitated by specialization, decentralization and cooperation, as well as the impact of public health, healthcare systems undergo a significant paradigm change moving from organization-centered care through process-controlled or shared care to personal health to improve safety and quality of care as well as efficiency and efficacy of care processes [1], [2]. The molecular analysis of the humans' structural constituents contributes to the understanding of disease pathogenesis and unraveled substantial Ínter-individual differences leading to diverse phenotypes of a given disease. Based on those differences and the stability of the molecular characteristics of a disease, promising therapies tailored to the needs of a particular patient can be derived. Such

personalized medicine opens the door to different fields of medicine. Thereby, the consideration of medical, but also socio-economic, legal and ethical aspects of this important move is inevitable, which will be addressed in more detail in [3]. The European Unión Priority Settings defmed personal health systems as its 2009 scope [4]

and established a road map for personal health systems and research themes for the 7^^' Framework Programme and beyond (PHS 2020), additionally including independent living - also known in the context of Ambient Assisted Living (AAL) [5]. This chapter introduces domain-specific, organizational, and technical paradigms, requirements and solutions for personalized. ubiquitous, care. Here we emphasize the Corresponding Author. Jacqueline Hall, BSc, PhD, Head of Translational Research Unit; EORTC European Organisation for Research and Treatment of Cáncer, Avenue E. Mounier83/11, 1200 Brussels, Belgium; Email: [email protected]; Phone: +32-2-774 16 58.

JA. Hall and B. Blobel/ Paradigm Changesin Health Lead to Paradigm Changesin Pathology

39

formal aspects of modeling and implementing Telehealthand personal health (pHealth) interoperability and the entailed multidisciplinary integration, and illustrate the drivers behind and benefits of personalized medicine with a specific focus on the changing trends and impact on pathology, especially emphasizing Telepathology.

1. Material and Methods

The advanced Telehealth paradigm, and its specialization to pHealth, is characterized by the patient's active involvement. The technical requirements for remotely enabling personalized health services, as well as a generic reference architecture model, will be shortly introduced. Furthermore, semiotic and knowledge representation aspects in the Telehealth context will be discussed. As a conclusión of this section, a formal and

unified methodology for Telehealth/pHealth system analysis and design based on an overall architectural framework has been developed and deployed for achieving the business objectives to provide high quality, personalized, ubiquitous care. Telehealth comprises health services provided to the individual independent of the time these services are delivered as well as the physical location of actors, services and resources (furthermore summarized as actors) involved in the care process. In other words, Telehealth is distributed care, i.e. health related interactions that connect

comprehensively communicating and cooperating actors over distance. Some of those interactions are mediated through technology in order to enable the appropriate and intended collaboration between the actors [2]. pHealth is defmed through the subject of care's status, needs, wishes and intentions and results in 'a form of medicine that uses

information about a person's genes, proteins, and environment to prevent, diagnose, and treat disease' which results in individually tailored diagnosis and therapy' [6].

Being in the center of, and controlling, the health business, the subject of care gets empowered. As a result, the care environment is a virtual organization with dynamic policies, workflows, technical as well as environmental conditions. Interactions and processes are dynamically created and performed, integrating medicine, informatics, biomedical engineering, information and communication technologies, bioinformatics, and the cluster of "omics" disciplines such as genomics, etc. Relations to epidemiology and public health are essential [7], [8], 9].

1.1. Technical Paradigmsfor Enabling Ubiquitous Individualized Health

For ubiquitous pHealth, i.e. individualized health services provided at any time to any location, three technological paradigms have to be implemented: mobile computing, pervasive computing and autonomic computing. Mobile computing enables the permanent accessibility of the principáis (persons, organizations, devices, applications, components) involved in providing, for example, teleconsultation services. Pervasive computing allows for remote and pervasive service provision through any type of principáis, thereby directly integrating the subject of care in the communication and cooperation network using sensors, actuators, body area networks [10], or home networks. These services are commonly called Telemedicine services. For providing personalized care or personalized medicine to empowered subjects, services have to be flexible and cannot be extemally and rigidly predefmed. Such adaptive health information system (HIS) design, which is directed towards a self-organizing

40

J.A. Hall and B. Blobel/ Paradigm Changes in Health Lead to Paradigm Changes in Pathology

environment, draws on current autonomic computing research and development challenges [11]. Together, ubiquitous care delivery is installed. 1.2. System-Theoretical, Architecture-Centric Approach to pHealth

For designing and implementing pHealth systems, first an agreement is needed on business objectives, business process, and involved entities. Second, we have to describe the problem space as well as the solution layout. To succeed on that matter,

the traditional and currently re-invented systems theory offers appropriate engineering methodologies for system analysis and design. Abstracting from all specialized aspects, a system of interest is defmed comprising of abstract objects (system components) of interest against its environment (the other objects), which interacts with that environment. The interest originates from certain structural and behavioural aspects the considered real world system with its components - the system's architecture - offers to realize the intended business process [1]. The recursively defmed system forming super-systems or sub-systems can be described

•

by its architectural components at different granularity (decomposition for analysis) or complexity (composition for design) levels (business concepts, relations network, aggregations, and details) [1], so modelling the system's structure,

•

and by the components' functions (operations) and interrelations realizing interactions, so modelling the system's behaviour. The representation of the described system's architecture is provided by the underlying ontology - for the abstract system by the general ontology. Different domains' experts consider different aspects of that system resulting in domain specific models. The system-theoretical, architectural approach works with any system such as living, technical, social, legal, etc. As those different aspects are combined in the real world system with its architecture, domain models reflecting the real world system's architecture provide the needed commonality despite the different concept space and the resulting representation of the domain specific architectural model by the corresponding domain ontology. In other words, it is not sufficient to care of the correctness of ontologies derived within the hierarchical schema of ontology systems. It is inevitable to verify them by the architecture and its representation through the general ontology of reality [12]. This fact is mostly ignored, resulting in wrong system models and ontologies (see, e.g., [13]). 1.3. The GCMas an Architectural Framework

The Generic Component Model provides a system theoretical architectural approach to any real world system [1]. It describes the architectural dimensión of component composition/decomposition of a system. The different granularity levels are interrelated through inheritance and constraints of the components. Because of its recursive character and the aforementioned four generic granularity levels, any real world system from elementary particle to the universe can be considered, so meeting the challenge of translational medicine [1]. By the GCM domain dimensión, any domain specific aspect can be modelled. If a system is considered at the subsystem level, thereby reducing its complexity, the domain will be specialized into sub-domains represented by sub-ontologies (branches in the ontology tree). The interrelation of components in the GCM is strictly restricted to the prevailing architectural level within

J.A. Hall and B. Blobel/ Paradigm Changes in Health Lead to Paradigm Changes in Pathology

41

a domain (component aggregation) and between domains (component binding). As the modelled solutions should be implemented, the GCM is completed by the development process dimensión according to the RM-ODP views.

2. Results

The advanced health paradigm, resulting from the shift towards Telemedicine and pHealth, is based on comprehensive communication and cooperation depending on knowledge representation harmonization up to the level of different languages, from that of patients to the one of health professionals. So, harmonization of terminology and ontology are crucial for the aforementioned paradigm changes. In this section we outline the challenges of knowledge representation, terminology and

ontology and give an example of personalized medicine as an illustration of the driving forces behind and the benefits of applying Telemedicine and pHealth. Lastly, we

present how this paradigm change of personalized medicine will also entail a cultural shift within health organizations with a specific focus of the impact on pathology. 2.1. The Knowledge Representation, Terminology and Ontology Challenge

As a subsystem is a system's component at a certain granularity level specialized by inheritance and constraints, the related ontology representing it is specialized according

to the hierarchy of ontologies from top-level, domain/subdomain, application down to application component, deploying inheritance and constraints within the related type hierarchy. The aggregation of components has to be performed at the appropriate level of architectural granularity for realizing the required ftinctionality. The binding to components belonging to other domains can only be performed at that level of granularity. Therefore, the interrelated ontologies as the representation of the architecture have to be specialized to the level of granularity required for bridging them. Components and relations of the ontologies have to be organized correspondingly. In other words, as the real world system's architecture is just described by the related system of ontologies, ontologies have to be designed and managed according to that architecture. A practical example of such an architecturebased ontology harmonization has been performed for automatically bridging between HL7 V2 and HL7 V3 using a mediating communication standards ontology [14]. Ontology coordination deals with the defmition of correspondences. Correspondences of systems and their components can be defmed according to their structure and behavior. So, we can architecturally define structural, operational and

relational correspondences. The presented multi-domain architectural model of a system represented by a set of ontologies offers a formal way to ontology harmonization. There are four ftindamental ways of ontology harmonization: matching,

mapping, alignment, and merging. In that context, different ways for investigating correspondences architecturally and ontologically (representation-related) are known [15]:

• • •

Component/concept ñames and descriptions Class hierarchies with super- and subclasses Defínitions of properties (ranges, areas, constraints)

•

Instances of classes

42

J.A. Hall and B. Blobel/ Paradigm Changes in Healtli Lead to Paradigm Changes in Pathology

•

Formal class description

2.2. The Drivers behind the Move to Personalized Medicine

Due to improvements in patient treatment in the last decades, there is a rise in the

number of patients suffering from chronic diseases that require long term management. Also many countries now have an aging population. These factors combined are leading to an increased number of individuáis requiring healthcare services [16]. For example, in the Netherlands, the ratio of elderly to productive adults is expected to increase to 30% in 2020 and to 46% in 2040 [17]. China also will face similar problems due to the adoption of the one-child policy in 1979. In tándem, recent innovation in technology combined with the increasing influence of the Internet [18], has allowed an increased throughput and lowered the cost of generating data, massively increasing our ability to perfonn molecular and genetic characterization and our ability to access information. Although further challenges still must be met in order to better interpret the mass of results and to fully intégrate these technologies into healthcare decisions in a quality assured way [19], mass data generation is leading to a better understanding of disease and has an increasing influence on healthcare decisions.

Lastly, with the movement to pHealth, the power of the individual is increasingly impacting on healthcare as patients expect a focus on quality outcomes and a customized and streamlined service taking into account personal preferences. These changing expectations coupled with a push towards increased choice for patients, will put pressure on organizations and governments to turn to 'proactive healthcare management'. Therefore, governments, patients and healthcare providers cannot afford to continué with outdated healthcare systems to meet these increasing needs. 2.3. The Benefits ofPersonalized Medicine

Personalized medicine can be broadly applied to two main areas, for preventative care and for tailoring therapeutic approaches once a disease has presented. In both cases, signifícant reduction in the cost and demand on healthcare systems can be achieved whilst providing better quality of care for patients [20]. In the area of therapeutic applications, targeting the right drug to the right patient can help avoid side-effects. This not only achieves the best outcome for the individual but will also reduce the

overall healthcare costs due to unnecessary treatment of patients who are not likely to benefit from the given therapy and to avoid the costs associated with harmful side effects.

Together, these forces are driving medical systems towards the holistic approach of personalized medicine, which offers the promise of providing effícient solutions for the increasing demand and need to deliver quality, comprehensive, personalized healthcare to patients. As a result of this patient-focused approach, relationships now start to revolve around the individual, changing the role of various partners involved in healthcare and their interactions. The new paradigm of personalized medicine will push health organizations to intégrate people and technologies and partner with other organizations, which have up until now, not been part of their typical routine.

J.A. Hall and B. Blobel / Pamdigm Changes in Health Lead to Paradigm Changes in Pathology

43

2.4. Key Paradigm Changes and Valué Creation in Pathology

Pathology has been on the brink of change for almost a decade. However, it is only in recent years that these changes begin to be fiilly integrated. Important changes in the role and tasks of the pathologist and the way that pathology departments are being organized are taking place (Figure 1). With tissue-derived molecular data, genetic information and images being combined with the patient's medical history to achieve effective treatments for the individual, pathology is faced with new challenges and this technology-driven demand for pathology services has sharply increased [21]. Pathology departments will need to be able to respond to these challenges by undergoing a cultural shift, involving changes in govemance processes and stakeholder interactions in order to realize its ftill potential as an essential diagnostic service in the new healthcare model. More than ever there is a need for clear leadership in this field, to help ensure that the change is well managed and results in tangible benefits.

Technoiogy

Legal

'>/

Frameworks

Figure 1. Changing Paradigms in Pathology.

Molecular profiling and the remaining importance ofhistology Although new technologies are leading to large volumes of molecular data allowing molecular characterization of diseases, the importance of morphology and simple staining techniques still remains. Some of the most important prognostic information may still be obtained from simple hematoxylin and eosin (H&E) staining with macroscopic description [22]. However, molecular data can also be used to advance molecular pathology and to better characterize diseases at the cellular level. For example sub-typing diseases using gene expression profiles has been very successful in breast cáncer to assess the molecular phenotypic heterogeneity [23]. More traditional techniques, such as immunohistochemical staining, also provide additional information on the cellular location and distribution of biomarkers such as expression on the cell surface membrane, in the nucleus or if certain proteins co-localize. This type of cell architectural information is not often obtained from genetic techniques and is critical to know as this can alter the accessibility of potential drug targets, or may change the interpretation of molecular assay results. Therefore, it is highly critical that different data types, including both molecular and histological information, are integrated.

44

JA. Hall and B. Blobel / Pamdigm Changes in Health Lead to Paradigm Changes in Pathology

2.5. The Importance ofBiobanking The identification of new diagnostic tools mainly depends on the correlation between genotype, phenotype and outcome. In that context, access to high quality, clinically well characterized samples will be the bottleneck in clinical research and development. A key element in the process towards personalized healthcare is the establishment of biobanks allowing population-wide investigations of genetic predispositions with a high statistical power. 2.5.1. Tissue biospecimensfor molecular testing

To allow advances in clinical research and to enable personalized treatments using companion diagnostics and histological information, tissue biospecimens will need to be analyzed. Thus, a key role for pathologists in the future will be to handle and store biospecimens, whereas in the past most pathology departments did not routinely have tissue or tumor banks (particularly of frozen tissue). With the move to personalized medicine, collection, storage and use of biospecimens will be one of the key tasks of the pathologist. Processes for the logistics will need to be ensured as this is essential for obtaining quality histological feature and biomarker analysis. 2.5.2. Quality assurance oftissue biospecimens

Histological analysis of tissues remains critical for assessing the heterogeneity of tissue biospecimens prior to them being used in diagnostic assays. As tissue specimens are composed of heterogeneous populations of cells e.g. stromal elements, fibroblasts, vasculature, nervous supply, infiltrating leukocytes, and ECM, and adjacent normal tissues in variable amounts, the portion of the sample contributing to the molecular analyses should be well characterized and controlled [24]. This has been proven to be extremely important for the adequate understanding of genomic data, since low tumor content vs. high tumor content may give very different results of high throughput genomic testing [25]. Also, the presence or absence of specific tumor markers depends on the sensitivity of the technique used, which in turn depends on the tumor cell content. Henee, adequate biospecimen characterization and handling according to harmonized laboratory standards is a major determining element of the success of subsequent molecular assays and the pathologist will play a key role in setting the standards for quality assurance and tissue characterization. 2.5.3. The right tissue biospecimen, in the right amount

The amount of genetic information obtained from tissue biospecimens is directly related to how the biospecimens have been treated and whether they were handled appropriately under the supervisión of a trained pathologist. For example, obtaining the mutational status of the tumor using high throughput genomic technologies requires a certain amount of DNA, and henee a certain amount of tissue. If more detailed

information is needed, such as identifying whether there are existing resistant tumor clones in the primary tumor, this will likely require a technique with a much higher coverage. Therefore, more DNA will be required and more tissue specimen is needed. The type of tissue and the manner of its preservation e.g. frozen tissue or formalin fixed material, will also impact on the amount and quality of genetic material that may be

JA. Hall and B. Blobel / Paradigm Changes in Health Lead to Paradigm Changes in Pathology

45

obtained from it and henee the range and number of biomarker tests that may be eonducted from it. Henee, the manner by whieh biospecimens are obtained from

patients should be related to the amount and type of information requested; a process in whieh the pathologist can play a key role.

2.6. Companion Diagnostics and Qiiality Assiirance ofMolecular Techniques The customization of healthcare has resulted in the development of targeted therapies

that act specifically in known sub-populations of disease that are identified by key molecular features, e.g. by single biomarkers or by complex multi-biomarker tests. Thus, companion diagnostic tests are needed to robustly identify the correct patient population for a given treatment, making these tests a core part of the overall treatment regime. Examples of applying this approach can be found in oncology were therapies are selected according to the genetic profiles of tumors and biomarkers now form part of national and intemational recommendations for the evaluation of colon, lung and

breast cáncer amongst others [26], [27], [28]. Without robust companion diagnostics, the effectiveness of the associated targeted therapies will be decreased or even lost. Therefore, the need for robust resuhs from molecular diagnostics is absolutely essential for the test to be useful in diagnostic and treatment settings. Harmonization in working practices and transparency through sharing and comparing test results between laboratories in external quality assurance rounds, as well as longitudinally over time within the same laboratory, will become increasingly important. Particularly in the case of multi-centre networks and pervasive healthcare. All this should be coupled with harmonized pathology reporting.

2.7. Pathology in Translational Research Translational research can be defmed as the bi-directional process of applying clinical

knowledge to basic research and scientific knowledge to clinical development. The benefits of translational research are shorter clinical development timelines and reduced risk by enabling earlier and more robust science-based decisión making. Pathologists will play an increasingly important role in innovation in healthcare through research and development, such as supporting scientific study and clinical trial design to allow translation of biomarkers and aid drug development as well as for consultation. Information from different sources (histology, staining, molecular data)

must be integrated to form a complete and holistic view allowing an accurate interpretation of the disease phenotype. As pathologists will be contributing valuable and unique information regarding the underlying pathology and molecular characteristics of the disease, they will be an important member of the multidisciplinary healthcare team. Pathologists will give a unique intellectual contribution and view that will need to be incorporated and integrated into the overall assessment of treatment options for the patient. 2.8. Legal Frameworks

Underpinning many of the aforementioned issues will be the applicable local legal framework and delineation of the custodianship of biospecimens. This could lead to

significant changes in the way that pathology and health research is eonducted at a practical level and will be reflected in the architecture of the system employed.

46

J.A. Hall and B. Bíobel / Paradigm Changes iti Health Lead to Paradigm Changes in Pathology

particularly since research and development may require large international and multicenter networks, information exchange and access to biospecimens across borders. 2.9. Reimbiirsement Policies

Reimbursement policies will impact organization and functioning of pathology laboratories. As companion diagnostics will be closely linked to specific treatments, reimbursement for pathology labs performing for diagnostic testing while reimbursing new related treatments will become important. Also, due to economic reasons, the implementation of govemment-imposed quality requirements will likely lead to the merger of smaller laboratories enabling the upgrade and maintenance of adequate facilities.

2.10. Information Technology (IT) for Supporting New Pathology Paradigms As emphasized at the start of this chapter, IT systems built on appropriate reference architecture models and technical requirements will be a powerfiil tool that will enable the discussed paradigm changes. Often, the success of navigating such changes will depend on the ability to develop efficient and suitable IT infrastructure that is user friendly enough to be easily integrated into daily practice. However, the development and use of IT does not come without challenges; the time and resources required to establish a robust and well organized IT system should not be underestimated. The applications of IT in pathology can range from digital pathology including scanning and digitization of slide images [29] up to automated recognition of cell phenotypes, automated scoring and automated image analysis, machine leaming and pattem recognition for generating IT-based expert rules for decisión support [30] as well as database systems for capturing, storing and sharing molecular, clinical data and images.

IT-based protocols will ensure tests are clinically appropriate via: • Electronic data capture: User friendly IT systems will simplify data capture, decrease workload and increase quality through efficient electronic data •

•

capture by both remote and on-site users. Using IT workflows and technology for automation of tasks: In the new environment a faster turnaround of specialized laboratory results will be key for allow efficient patient diagnosis, treatment and discharge. Automation of laboratory tasks (e.g. decapping, recapping, centrifugation and aliquotting) and integration into IT protocols and workflows will help meet this demand. This will also mean that skilled staff trained for operating automated platforms will be needed. To fully reap the benefits from automation and IT infrastructure, signifícant fmancial investment will be essential. IT interoperability: enabling networks of communicating clinicians and patients.

•

Supporting quality in biobanking: Digitized pathology images can be used for cataloging and characterizing tissue collections, recording histology, allowing easy referencing of older cases and selection of the correct cases needed for new research projects.

•

Supporting infrastructure: Storage, maintenance and backup of large databases of digitized images, molecular and clinical data will be fundamental to

J.A. Hall and B. BlobeU Paradigm Changes in Health Lead to Paradigm Changes in Pathology

47

maintaining quality records and capturing the information generated about a particular patient. Such databases must be integrated to allow information of various sources to be used and cross-referenced enable a holistic view. 2.11. Networks

The recruitment of population-wide cohorts, the inclusión of multi-disciplinary resources throughout the field of medicine and natural sciences, the consideration of

exogenous factors such as nutrigenomics, food safety and toxicology, the required information technology, requires international and discipline-crossing collaboration. Auffray et al. recommend the development of systems medicine through an international network of systems biology and medicine centers dedicated to interdisciplinary training and education, pushing the evolution of the presented paradigm change in developed countries, at the same time avoiding a digital divide to developing countries [31], [32]. In that context, the new structure of stakeholders to be involved in

the change in pathology has to be mentioned. Especially the empowered patient - or even better the citizen before becoming a patient - deserves a key attention. Table 1 ñames the major stakeholders to be considered. Table 1. A list of the major stakeholders in the new healthcare paradigm changes affecting pathology.

1. Patients (individuáis taking increased responsibility for personal wellness, stronger consumer preferences);

2.

Clinicians providing treatment recommendations on the basis of information generated and provided by pathologists;

3. 4.

Molecular biologists/technologists enabling high throughput molecular testing in a quality assured manner, providing tools for molecular pathology; IT/ computer experts enabling capture and use of molecular and digital data;

5.

data warehousing, infrastructure, eHealth; Goveming bodies and regulators including hospital boards, institution boards,

national govemments managing the co-ordination of interdependent parts of the health system, encouragement of partnerships, payment and fínancing, regulatory reforms that reward competition and innovation, addressing behavioral, genetic and medical system influencers; 6.

Pathologists moving from an observer / describer role to someone who can suggest modalities of personahzed therapy based on the fiinctional properties of cells and tissues.

3. Discussion

According to [31], the exploration of structure, variation and function of the human

genome through high-throughput technologies for DNA sequencing and for analyses of transcriptomes, proteomes and metabolomes enable hypotheses and predictions about health and disease states. Furthermore, it provides possibilities for analyzing a large number of individual genomes and transcriptomes. In consequence, complete reference

proteomes and metabolomes can be provided using powerful analytical techniques based on chromatography, mass spectrometry and nuclear magnetic resonance.

Computational and mathematical tools support the understanding of functional and

48

J.A. Hall and B. Blobel/ Paradigm Changes in Health Lead to Paradigm Changes in Pathology

regulatory networks ruling the behavior of complex biological systems. The advancement of those tools enables the integration of various data types across the complexity levels and time frames of biological systems, which characterize human development, physiology and disease [31]. The described approach provides better understanding of the mechanisms of human diseases and facilitates the development of better diagnostic and prognostic biomarkers for cáncer and many other diseases. Based on closer collaboration between science and industry, the development of drugs will change, targeting múltiple components of networks and pathways impacted by diseases [31].

Based on the mass of data about the human system and its diseases and resulting mathematical and computational models, systems medicine enables individualized preventive and predictive care.

4. Summary

The new healthcare paradigm involves a complex array of interactions and interdependent parts of the health system. Pathologists will assume a much more active role in the management of diseases and will be a key part of inter-disciplinary teams. Education of the various stakeholders towards the changing role of pathology and the development of different economic laboratory models depending on the purpose of the laboratory and what pathologists can offer in the new health paradigms is also important. Increasingly, it is the patient who will own the test result as patients gain direct access to the full range of test results that have been conducted. Pathologists will need to take on management roles and champion quality assurance, legal issues, negotiation and interaction with múltiple stakeholders, awareness of fmancial issues, budgeting and regulatory matters. These changing responsibilities will need to be reflected by updated career frameworks and appropriate education for pathologists, to build and maintain the knowledge, skills, competences and attitudes required. This will need to be combined with appropriate workforce planning to ensure a continuous supply of trainees. The results will yield a more nimble pathology service, enhance quality, reduce risks, and increase efficiency. This will rest on the foundations of sound evidence based medicine and of excellent enabling technologies.

Acknowledgement

The authors are indebted to the colleagues of the COST IC0604 Action as well as to the SDO activists from HL7 Intemational and IHTSDO for kind collaboration and

excellent support.