Introduction
Autism Spectrum Disorders (ASD) is an umbrella term that covers a multitude of early-onset neurodevelopmental conditions characterized by a wide array of cognitive, behavioral, and social impairments [1]. The symptoms exhibited by ASD patients are diverse, as ASD is very heterogeneous in its manifestation, with a 4:1 male to female gender ratio and a wide range of symptom severity ranging from highly functioning individuals who can lead a relatively independent life as adults to severely mentally impaired individuals who would require constant caretaking and supervision throughout their lives [2], [3]. Despite the diversity of symptoms, however, most ASD individuals suffer from a reduced quality of life, as even high functioning individuals still face social challenges, which can be a source of distress; furthermore, ASD also incurs a tremendous emotional and financial toll on the family of ASD individuals as well as society as a whole [4].
Research on ASD has progressed rapidly over the past two decades [5]. ASD research, however, is extremely multi-faceted, ranging from basic scientists seeking to understand the molecular and cellular underpinnings of the disorder through utilizing animal models and molecular neurobiology approaches to behavioral scientists working with patients in order to develop new approaches to help the parents and educators better take care of individuals afflicted with this disorder. This diversity in research makes it particularly challenging to bridge the gap between such disparate disciplines; furthermore, as the research environment grows increasingly complex, the development of techniques ensuring the accessibility of this research output to practicing physicians and primary caretakers becomes of paramount importance [6], [7], [8].
Biomedical ontologies have recently witnessed a remarkable increase in popularity as the tool of choice for bridging interdisciplinary gaps and ensuring the widespread accessibility and exchange of information between researchers and professionals with diverse backgrounds and a broad range of specializations [6], [7], [8], [9], [10], [11], [12], [13], [14]. The striking success of biomedical ontologies stems from their ability to provide a comprehensive framework through which investigators can share their findings and where professionals can easily access the studies upon which their diagnostic and treatment approaches are based.
While several biomedical ontologies have been made for autism, they suffer from their focus on specific aspects of ASD diagnostic and risk factor information; therefore, their uses are limited in the scope of their potential applications [6], [7], [8]. In order to address the limitations of the currently available autism ontologies, we sought to develop a universal autism ontology (AutismOnt) that overcomes the existing limitations through exhaustively covering all aspects related to autism researchers and primary caretakers. The expanded hierarchical organization of AutismOnt, makes it one of the most reliable and versatile ontology for the development of future text annotation and decision support systems for ASD researchers, physicians, and caretakers.
Ontologies are used in the medical domain mainly for representing and reorganizing medical terms and terminologies. Building medical ontologies is a challenging task that requires a deep analysis in the medical domain; nevertheless, this analysis is the core of building the medical ontologies that handle issues in the medical field. In this study, the main contributions of AutismOnt ontology are:
• Developing a universal autism ontology that holds a large amount of the most useful autism knowledge from various sources and covers different autism research directions.
• Conceptualizing the available knowledge in different aspects of autism and enabling it to be shared and used by the clinical and research community.
• Providing a comprehensive semantic map for domain terms.
The autism ontology developed in this study unifies some domain terminologies and help practicing physicians and primary care providers in their diagnosis and treatment decisions, which will reflect positively on the autistic individuals' lives.
The rest of the paper is organized as follows: Section 2, summarizes related work. Section 3, methodology is explained. Section 4 the AutismOnt ontology is proposed. Finally, we conclude by pointing out some future research lines.