Graph based clinical decision support system using ontological framework

Abstract

Scarcity of doctors in rural areas of developing countries is a major problem and has serious impact in health sector of villages. Health kiosks driven by the health assistants in different remote places are the backbone of rural healthcare services. However, due to limited knowledge and experience of the health assistants, diagnosis is often ambiguous. Therefore, there is an increasing demand to develop a knowledge based decision-making system to treat the rural patients at primary level. In this paper, a graph based clinical decision support system (CDSS) has been proposed to facilitate the health assistants for provisional disease diagnosis of the patients. The graph-based knowledge base is developed by integrating the medical knowledge represented of different ontologies. We apply the modified depth first search algorithm and topological sort algorithm for achieving minimum cost in graph traversal for differential diagnosis of the diseases. Diagnosis may be performed in two modes – online and offline, in the presence of the patient and using patient records respectively.