A Processing Model Using Natural Language Processing (NLP) For Narrative Text Of Medical Record For Producing Symptoms Of Mental Disorders

Abstract

Mental health is still a serious problem in Indonesia. Data on basic health research in 2013 showed the prevalence of serious mental disorders in Indonesia reached 1.7 permil. The ratio of professional psychiatrists with the number of sufferers is 0.47 per 100000 patients which means that they still do not meet the requirements of WHO. This has an impact that not all people with mental disorders can be handled by professional psychologists directly. Therefore, the development of a case-based reasoning system (CBR) to help diagnosing types of mental disorders can be an option to overcome these gaps. In most health care places, an initial examination of mental patients is carried out by medical professionals who are not experts. At a certain level, the patient cannot express the symptoms that are felt, so the examiner can make narrative texts about the patient's condition. On the other hand, text documents are not structured that are commonly processed in CBR systems, so they cannot be directly applied. Therefore a text processing model to become symptoms is required. In this paper, the text medical record processing model using Natural Language Processing (NLP) is discussed to produce symptoms of mental disorders. This model is a part of the development of a case-based reasoning computer system to help diagnosing the types of mental disorders and their management.