Ontology based Disease Diagnosis using Natural Language Processing, SPARQL and Protégé from Patient Symptoms

As medical science is evolving; more diseases are being detected and have forayed into our world with high mortality rates. Accurate and immediate diagnosis and treatment has become extremely important. Some diseases take time to show distinctive symptoms but decline in health is extremely fast. This paper proposes a way of assisting individuals having marginalized access to healthcare and aims to decrease the ever-growing pressure on doctors. We aim to develop a model that will be able to give a classification of the disease that the patient might be suffering from after inputting the patient's symptoms in English or their native language using voice commands or text. Ontology has been created on the software Protégé, which contains information of about 13000 diseases out of which almost 5000 are infectious. Here, the causal insights will provide the various relationships between symptoms, symptom hierarchies, diseases and disease super classes. While this information might not be useful for the patient, the same will assist the doctor in providing faster treatment. Hence, this model can also be used by doctors for reconfirmation of their diagnosis and to revalidate the treatment being provided after generating the knowledge graph.