Effective learning and knowledge discovery using processed medical incident reports

Effective management of technology plays an essential role in defining the power of an arena. In many developed countries, such as Japan, healthcare facilities employ advanced information systems to capture daily healthcare records. We have collected thousands of incidence reports from the Japan Council for Quality Health Care, which is managed by the Ministry of Health, Labour and Welfare. The incident reports were electronically stored in written conversation format. We successfully distinguished the incident reports using artificial intelligence technology. Using natural language processers, Japanese vocabularies were systematically structured, captured and classified. As a preliminary, we explored the similarities between reports and the co-occurrence events of related characters among medical incidences. In this study, we took advantage of advanced health informatics approaches and available encrypted datasets to extract hidden knowledge associated with medical error events. The occurrence of medical errors, such as inappropriate oral medicine, may be statistically associated with and can be explained by event scenes. This data-driven research involves the intimate collaboration and technology management of statisticians, computer scientists, and practitioners - a concept known as “Convergence” and attempts to statistically understand the dynamics of medical incidences to enhance clinical patient safety situation awareness.