Opportunities and challenges for applying process mining in healthcare: a systematic mapping study

Abstract

Process mining applies robust methodologies using data mining and machine learning for pattern recognition, using models that represent the process flow identified by the sequence of events, their timing, and the assessment of resources used. To evaluate the use of process mining in health care, with emphasis on the identification of characteristics, health care studies were selected based on a systematic review of the literature, well-defined eligibility criteria, and guided research questions. Such questions address the strategy and algorithm adopted, the location used, and the main contributions for the identified application. A total of 270 articles were selected. Among the identified applications, the discovery of process models was the most frequent, followed by resource analysis and evaluation. The most adopted algorithms were identified, the Fuzzy Miner and Heuristic Miner. One may highlight, among the main contributions, the analysis and discovery of process models for the evaluation of patient care and the evaluation of process conformity, focused on medical protocols and clinical guidelines. This review highlighted the significant use of process models discovery in their evaluation, thus supporting the proposal of changing the health care model so that it favors resources evaluation and care quality. There is also an important challenge regarding the use of such technique; on the one hand, concerning data integration and a more automatic recognition of standards and, on the other hand, concerning the application of standards focused on needs for compliance evaluation between discovered models, medical protocols and clinical guidelines.