Using Diagnostic Information to Develop a Machine Learning Application for the Effective Screening of Autism Spectrum Di
A 2-Class Support Vector Machine (SVM) classification model was developed by means of machine learning techniques and text analysis of Autism Spectrum Disorders (ASD) diagnostic reports. The ability of the 2-Class SVM application to screen for ASD is comp
- PDF / 331,603 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 104 Downloads / 142 Views
Using Diagnostic Information to Develop a Machine Learning Application for the Effective Screening of Autism Spectrum Disorders Tze Jui Goh, Joachim Diederich, Insu Song and Min Sung
13.1 Introduction Recent studies have found an increase in the prevalence of Autism Spectrum Disorders [7, 22, 53]. In the United States, the [6] estimated the overall prevalence rate to be 9 per 1,000. In the United Kingdom, a rate of 11.6 per 1,000 was found by, [2], of which childhood autism was estimated to range between 2.48 to 3.89 per 1,000. These figures emphasize the pertinence of ASD, especially when the impairments associated with the disorder are all-encompassing and impact the quality of life of individuals with ASD and their families. The increasing numbers can be attributed to greater awareness of the disorder, thereby resulting in reclassification of previously ‘mis-diagnosed’ cases and also improved efficiency in identifying individuals with ASD. A broadening definition of ASD, i.e., inclusion of Asperger Disorder and Pervasive Developmental Disorders-Not Otherwise Specified (PDD-NOS), in addition to Autism in the Diagnostic and Statistical Manual (DSM-IV, 1994) and International Statistical Classification of Diseases and Related Health Problems—10th edition (ICD-10) [63] is also a fundamental contributing factor [59].
T. J. Goh (&) M. Sung Institute of Mental Health, Singapore, Singapore e-mail: [email protected] M. Sung e-mail: [email protected] J. Diederich School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane Q4072, Australia e-mail: [email protected] I. Song School of Business and IT, James Cook University, Singapore Campus, Singapore 574421, Singapore e-mail: [email protected]
M. Lech et al. (eds.), Mental Health Informatics, Studies in Computational Intelligence 491, DOI: 10.1007/978-3-642-38550-6_13, Springer-Verlag Berlin Heidelberg 2014
229
230
T. J. Goh et al.
Autism is characterized by a triad of impairments [61] in the areas of reciprocal social interaction, communication, and repetitive and stereotyped behaviours. Asperger Disorder is similar to Autism but requires no deviance or delay in language development, according to the DSM-IV-TR (1994). PDD-NOS is diagnosed when there are significant impairments in the areas similar to Autism, but the individuals do not fulfil sufficient criteria for the diagnosis of such. Although the [18] and [33] guidelines are helpful to clinicians in the diagnostic process, the effectiveness of their utilization depends on the experience of the clinicians [38]. A brief examination of both traditional diagnostic systems reveals that the diagnostic criteria of the disorders are general descriptions of behavioural indicators which are open to the subjective interpretation, clinical observations and accuracy of reporting by informants [43]. The manifestation of symptoms also varies across different individuals and across different chronological and developmental levels [9], all of which form the spectrum natu
Data Loading...
