Chatbot Implementation for ICD-10 Recommendation System

Abstract

ICD-10 coding becomes an essential process to transform descriptions of medical diagnoses and procedures into universal medical code numbers. Because there are a large number of the ICD-10 codes, this process needs an expert or an experienced staff to proceed. However, insufficient of the staffs in this area makes this task become a difficulty for general public health staff. Therefore, a recommendation system by using chatbot technology is proposed in this work. This system is implemented by using a messaging application with auxiliary Natural Language Processing (NLP) library. The system was compared with the conventional ICD-10 application by using Analytic Hierarchy Process (AHP). The evaluation result shows that the proposed chatbot can perform an effective solution for selecting ICD-10 code. In addition, it also shows that the proposed chatbot is preferable to be applied in the ICD-10 application.