AWERProcedia Information Technology & Computer Science

2 downloads 902 Views 486KB Size Report
AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. ... Tehran, Iran, E-mail Address: ms[email protected] / Tel.
AWERProcedia Information Technology & Computer Science Vol 04 (2013) 1069-1073

3rd World Conference on Innovation and Computer Sciences 2013

Online expert system for screening autism: an item analysis Mohsen Shokoohi-Yekta *, Associate professor, University of Tehran, Chamran Hwy, Tehran, Iran. Maryam Mahmoudi, Ph.D student, Allameh Tabataba’ i University, Tehran, Iran. Bagher Ghobari-Bonab, Associate professor, University of Tehran, Chamran Hwy, Tehran, Iran. Hamid Reza Pouretemad, Associate professor, Shahid Beheshti University, Tehran, Iran. Salahadin Lotfi, Supervisor of Neurofeedback and Cognitive Training Section, Atieh NeuroscienceBased Center, Tehran, Iran. Suggested Citation: Shokoohi-Yekta, M., Mahmoudi, M., Ghobari-Bonab, B., Pouretemad, H., R. & Lotfi, S. Online expert system for screening autism: an item analysis. AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. Available from: www.awer-center.org/pitcs Received December 18, 2012; revised January 17, 2013; accepted March 28, 2013. Selection and peer review under responsibility of Prof. Dr. Fahrettin Sadıkoglu, Near East University. ©2013 Academic World Education & Research Center. All rights reserved. Abstract The number of children diagnosed with autism is growing rapidly and is a warning sign considered internationally. Autism is a neurological disorder that impedes communication, behavior and social interaction. This disorder has a crucial effect on individuals, functioning, as well as on their educational and social affairs. The aim of the present study was to develop an expert system to help professionals to screen autistic children accurately. It provides comprehensive information on autism helping to diagnose autistic children. To this end, this study is preliminary one that to select the questions required by the system. A questionnaire was developed based on the related literature, and then, has been studied by the experts in the area of psychology and special education, selecting more important questions on the basis of professional judgments. The final questionnaire was completed by 53 parents of autistic children in a center for the treatment of autistic disorders, 49 parents of normal children, and 20 parents of children with Down syndrome. The data were analyzed using Cronbach's alpha and Chisquare. Questions were selected on the basis of distinguishing Down Syndrome, and normal groups from autistic children. Finally, theoretical implications and practical applications of the test has been delineated. Keywords: First keywords, second keywords, third keywords, forth keywords;

*ADDRESSES FOR CORREPONDANCE: Mohsen Shokoohi-Yekta, Associate professor, University of Tehran, Chamran Hwy, Tehran, Iran, E-mail Address: [email protected] / Tel.: +98-21-6111-7452

Shokoohi-Yekta, M., Mahmoudi M., Ghobari-Bonab, B., Pouretemad, H., R. & Lotfi, S. Online expert system for screening autism: an item analysis AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. Available from: www.awercenter.org/pitcs

1. Introduction Autism spectrum disorder is a neurodevelopmental disorder that characterized by impairments in the social interaction, verbal and nonverbal communication and stereotypically and repetitive behaviors or interests (America Psychiatric Association, 2000). Several studies suggest an increased prevalence of autism; as in 1987, the prevalence of autism was from 4 to 5 per 10,000; and in 2007, was one in every 150 children in America (Gallo, 2010). In Iran, the prevalence of autism in the preschool screening for children five years old was nearly 24 out of every 10,000 children, although the prevalence of children who subsequently used the Diagnostic Interview for Autistic Disorder was approximately 6.26 per 10000 (Samadi, Mahmoodizadeh & McConkey, 2011). However, an accurate estimate of autism prevalence in children under five years old is not available. Delay in the diagnosis and treatment of autism may manifest to have a worse prognosis (Corsello, 2005). Early detection of autistic disorder is based on the combination of different strategies, such as developmental history through the assessment of communication, social skills, play and parental concerns (Filipek, et al., 2000) and despite medical advances, no biological markers exists for the diagnosis of Autistic Disorder, therefore, still diagnosis of the disorder is based on the detailed analysis of the behavior (Lord et al., 2000). There is a general agreement that the earlier the intervention begins, the better the effects have remained (Marteleto & Pedromonico, 2005). Progress towards early diagnosis has been slow due to the limitations in the available assessment tools for the diagnosis of autism (Dolan, 2009). Researchers have noted that the best method of diagnosis is a clinical judgment by experienced professionals (Charman & Baird, 2002). In conjunction with clinical judgment, there is not usually a consensus among professionals, so, families sometimes have nightmares and are confused by multiple diagnoses, and thus they may not start educational and clinical interventions on time. In contrast, Tools alone are not reliable in diagnosing autism. Available tools, are for young populations with poor psychometric characteristics, classification error and poor discriminant power (Gray, Tonge & Sweeney, 2008) and do not have sensitivity to cultural factors. In Iran, there is no scale designed for autistic children in lower ages, and that, some current tests may not have enough sensitivity and specificity. Thus, using expert systems, this shortcoming may be overcome. Expert systems are computer programs that can simulate the expert judgment and help problem solving. The system is based on the rules in the knowledge base of the program that is equivalent to the expert knowledge. The inference engine, which is another part of the system, has rules to conduct and simulate the expert's reasoning process (Turbon, Aronson & Liang, 2007). Finally, given these limit actions, the propose of the current study is a preliminary design of an expert system for diagnosing autistic children aging 2 to 6, focusing on the required items. 2. Method 2.1. Design and sample group. The design of this research is a descriptive-survey one. The study population included all of the boys and girls with autistic disorder aging 2 to 6 years in Tehran. The study sample size was 53 mothers of children with autism, 49 mothers of normal children and 20 mothers of children with Down syndrome (see table 1). An available sampling method was used for all of the three groups.

1070

Shokoohi-Yekta, M., Mahmoudi M., Ghobari-Bonab, B., Pouretemad, H., R. & Lotfi, S. Online expert system for screening autism: an item analysis AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. Available from: www.awercenter.org/pitcs

Table 1. Demographic information of three groups of autistic, Down syndrome and normal children Group Autism

Down Normal

Variable Current age Diagnostic age Mother’s education Final GARS score Current age

Mean 4.6 3.5 10.6

STD 7.07 6.02 39

Minimum 3 1.11 0

Maximum 1.2 1.4 10.9

78.7 6.1 4.5

115 7 6

55 6 2

15.4 0.4 1.5

2.2. Procedure This study aimed at examining psychometric properties of the items designed to for expert system and thus, began by evaluating the instrument developed for this purpose. To collect items for this instrument, the following data sources were used: 1- Diagnostic and Statistic Manual of Mental Disorders- Forth Edition-Revised, 2- comprehensive psychological textbooks, such as “Comprehensive textbook of psychiatry” (Sadock & sadock, 2009), The gale encyclopedia of mental health (Fundukian & Wilson, 2008) and The encyclopaedia of autism spectrum disorder (Turkington & Anan, 2007); 3available articles in the scientific information databases (such as Elsevier, Proquest and Springer); 4items from diagnostic questionnaires and interviews for autism disorder (such as Gilliam Autism Rating Scale, Social Communication Questionnaire) and interview with clinical psychologists specialized in autism diagnosis. After studying the required data, a set of items comprising of 481 items were collected. In the next step, the items have been studied for the repeating items, items representing more than one symptom or problem, eloquence of speech and writing, and thus, some items were deleted. Then, three experienced psychologists revised and screened the items. Finally, a total of 238 items remained. 3. Results The purpose of the item analysis was to study the psychometric properties, including validity and reliability, of the instrument. Since this tool was originally developed under the supervision of three psychology and psychometric professionals, thus, content validity was confirmed. Additionally, Cronbach's alpha coefficient (with and without deleting each item), and the correlation between each item score and the total score of each category was calculated. Based on the analysis conducted, the number of subscales reduced from 30 to 15 and the number of items fell from 238 to 167. The coefficients of internal consistency (Cronbach’s alpha) were calculated for each of the four categories representing the internal consistency of each category (see Table 2). Table 2. Internal consistency Coefficients of each category and number of items related to each subscale Categories

Social interaction

Internal consistency

Subscales

0.96

Impairment in nonverbal behaviors Failure to relation with peer Lack of sharing enjoyment with others Lack of social and emotional interaction

Number of items 18 17 22 10

1071

Shokoohi-Yekta, M., Mahmoudi M., Ghobari-Bonab, B., Pouretemad, H., R. & Lotfi, S. Online expert system for screening autism: an item analysis AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. Available from: www.awercenter.org/pitcs

Communication

0.94

Stereotyping behaviors

0.85

Developmental delay

0.85

Lack of theory of mind Delay in verbal communication Destroy in initiate relation to other Stereotyping language Lack of play Occupation Inflexible following from rituals Repetitive gesture delay Self helping skills Regulation factor

10 21 9 8 8 9 4 6 10 6 9

First category, social interaction (76 items & 5 subscales), with internal consistency of 0.96; second category, communication (47 items & 4 subscales) with internal consistency of 0.94; third category, stereotyping behaviors (19 items & 3 subscales) with internal consistency of 0.85 and forth category, developmental delay, (26 items & 3 subscales) with internal consistency of 0.85. As can be seen, categories in terms of internal consistency are in good condition. In other words, the items in each category are highly correlated. Correlation coefficients of each item with the total score for each category were determined. For social interaction category, correlation coefficients for each item with its total score was between 0.258 (item 6) and 0.723 (item 77); for category of communication, coefficients were between 0.21 (item 101) and 0.76 (item 137); stereotypical behaviors, between 0.2 (item 194) and 0.55 (item 175) and developmental delays between 0.26 (item 214) and 0.56 (item 232). These coefficients are discriminant indexes of the items, Therefore, They indicate of each item to detect interindividual differences. The minimum value of this index is considered to be 0.30; and the vast majority of indicators in the domains are higher than this amount. Only a few items were lower than 0.30, but because of their theoretical importance, they were not removed. 4. Discussion One of the reasons for the delay in the diagnosis of autism is the fact that current diagnostic criteria and measures identifying the autistic behaviors are based on the absence rather than emphasized the presence of specific indicators. This can lead to the misdiagnosis, and the absence of the same behaviors can be seen in many other developmental delays and disorders (Gray & Tonge, 2001). One advantage of this tool is that both kinds of behaviors, whether the behaviors that presence of them is a determinant of Autism or behaviors that lack of them indicates such disability. This study examined the psychometric characteristics of the items included in the expert system. Thus, almost items of this instrument were explained theoretically and the categories had high internal consistency. Categories also had good content validity. Underlying constructs are related to the factors differentiating autism form the other two groups. These items were also very comprehensive. We hope that the items provide an appropriate expert system to help diagnosing autistic disorders. The study had some limitations, including small sample size, which needs to be taken into consideration when interpreting such findings. So it can be suggested to the future researchers to conduct such studies with large sample sizes. Acknowledgement We are thankful for Dr. Hadi Moradi, Amir Ali Bagherzadeh, the mothers of autistic children and the stuff in the Center for the Treatment of Autistic Disorder (CTAD) because of their sincere collaboration that made this research possible. 1072

Shokoohi-Yekta, M., Mahmoudi M., Ghobari-Bonab, B., Pouretemad, H., R. & Lotfi, S. Online expert system for screening autism: an item analysis AWERProcedia Information Technology & Computer Science. [Online]. 2013, 04, pp 1069-1073. Available from: www.awercenter.org/pitcs

Reference American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th Ed.). Washington, DC: American Psychiatric Association. (Text revision). Charman, T., & Baird, G. (2002). Practitioner review: Diagnosis of autism spectrum disorders in 2 and 3 year old children. Journal of Child Psychology and Psychiatry, 43, 289–305. Corsello, C. (2005). Early intervention in autism. Infants and Young Children, 18, 74–85. Dolan, W. N. (2009). Using the autism diagnostic observation schedule (ADOS) to discriminate between children with autism and children with language impairments without autism. The Graduate Faculty of the Louisiana State University and Agricultural Mechanical College, Master of Arts in The Department of Communication Sciences and Disorders. Filipek, P. A., Accardo, P. J., Ashwal, S., Baranek, G. T., Cook, E. H., Jr., Dawson, G., & Volkmar, F. R. (2000). Practice parameter: Screening and diagnosis of autism. Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Child Neurology Society, St. Paul, MN. Gallo, D. P. (2010). Diagnosing autism spectrum disorders, a lifespan perspective. UK: Wiley-Blackwell. Gray, K. M., & Tonge, B. J. (2001). Are there early features of autism in infants and preschool children? Journal of Paediatrics and Child Health, 37, 221-226. Gray, K. M., Tonge, B. J., & Sweeney, D. J. (2008). Using the autism diagnostic interview-revised and the autism diagnostic observation schedule with young children with developmental delay: Evaluating diagnostic validity. Journal of Autism and Developmental Disorders, 38, 657–667. Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L., DiLavore, P. C. (2000). The autism diagnostic observation J Autism Dev Disord 123schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism of Developmental Disorders, 30, 205– 223. Marteleto, M. R. F., & Pedromonico, M. R. M. (2005). Validity of Autism Behavior Checklist (ABC): preliminary study, Rev Bras Psiquiatr, 27(4):295-301. Samadi, S. A., Mahmoodizadeh, A., & Mcconkey, R. (2011). A national study of the prevalence of autism among five-year-old children in Iran. SAGE Publications and The National Autistic Society, 1–13, 1362-3613. Turbon, E., Aronson, J. E., Liang, T. P. (2007). Decision Support Systems and Intelligent Systems. New jersey: Pearson.

1073