Applying Semantic Computing for Health Care Professionals: the Timing of Intervention is the Key for Successful Rehabilitation

Abstract

This study aims to apply Graph Machine Learning, a subset of artificial intelligence, in labeling electronic health records. The theoretical approach of the study stems from the studies of AI, machine learning, health policy, and physical medicine and rehabilitation. The data of chronic low back pain patients (n=93) were collected from electronic health records in form of a free text. The comparative analysis between the AI and medical expert was executed with the data of randomly selected patients (n=5). The International Classification of Functioning, Disability, and Health was used as a scientifical frame to identify the factors affected by a patients medical status. A medical expert identified the factors stated in the electronic health records. Data was analyzed and labeled with the graph (semantic networks) based machine learning engine, Headai Graphmind. Headai Graphmind automatically converted the findings to a readable map of factors, which are relevant concerning the timing of rehabilitation. Headai Graphmind found 56% of identical factors in relation to the medical expert. In future studies, the analyses of mutuality between Headai Graphmind, the health care professional and the patient are crucial to set the right timing for rehabilitation.

Keywords

• graph machine learning
• semantic computing
• headai graphmind
• health policy
• rehabilitation