An ontology-based approach for depression diagnosis

• Published in: AIP conference proceedings Vol. 2579; no. 1
• Main Authors: Nordin, Noratikah, Zainol, Zurinahni, Noor, Mohd Halim Mohd, Fong, Chan Lai, Buji, Ryna Imma
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 06.10.2023
• Subjects: Diagnosis; Knowledge based systems; Knowledge representation; Programming languages
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0136298

A high-risk population for suicide attempts was found from a depressed patient. However, most people with depression are not aware, untreated, and misdiagnosed. The presence of semantic inconsistencies in the knowledge-based system and limited symptoms in representing the depression domain knowledge due to varying from one person to one person lead to a major challenge in diagnosing depression. This paper proposes an ontology-based approach for depression diagnosis by collecting and studying depression knowledge of subjects for further diagnosis. All concepts and relationships detailed are applied using Web Ontology Language (OWL). The findings show that the proposed model can identify an individual with depression with 80% accuracy, 70% specificity, and 90% sensitivity. The proposed model can be used as a reference diagnosis model and the classification of an individual with depression will be improved by adding the information on risk factors that affect depression for further analysis.