Towards the Use of Standardized Terms in Clinical Case Studies for Process Mining in Healthcare

Abstract

Process mining can provide greater insight into medical treatment processes and organizational processes in healthcare. To enhance comparability between processes, the quality of the labelled-data is essential. A literature review of the clinical case studies by Rojas et al. in 2016 identified several common aspects for comparison, which include methodologies, algorithms or techniques, medical fields, and healthcare specialty. However, clinical aspects are not reported in a uniform way and do not follow a standard clinical coding scheme. Further, technical aspects such as details of the event log data are not always described. In this paper, we identified 38 clinically-relevant case studies of process mining in healthcare published from 2016 to 2018 that described the tools, algorithms and techniques utilized, and details on the event log data. We then correlated the clinical aspects of patient encounter environment, clinical specialty and medical diagnoses using the standard clinical coding schemes SNOMED CT and ICD-10. The potential outcomes of adopting a standard approach for describing event log data and classifying medical terminology using standard clinical coding schemes are further discussed. A checklist template for the reporting of case studies is provided in the Appendix A to the article.

Automatic and Explainable Labeling of Medical Event Logs With Autoencoding

“…As an example, the integration of free text with structured medical information to create pertinent medical labels could be interesting to consider in future works. As noted by Helm et al [28], a lack of sufficient coding in existing case studies remains. Therefore, addressing such complexities in the method will be interesting for practical uses on case studies with insufficiently coded data.…”

Process mining applications in the healthcare domain: A comprehensive review

“…Motivated by the fact that previous survey articles did not report clinical aspects in a uniform way, did not follow a standard clinical coding scheme, and details of the event log data were not always described, Helm et al (2020) surveyed 38 PM studies in healthcare, published between 2016 and 2018, with an emphasis on the details of the event log data, algorithms and techniques used. In particular, of the reviewed papers they described the characteristics of the event log, and referred to a standard clinical coding scheme for the information on clinical specialty and medical diagnoses.…”

Comparative Assessment of Data-driven Process Models in Health Information Technology

“…PM4CA in healthcare has been applied largely to both the analysis of healthcare processes and clinical guidelines. Recent studies include the following: Rinner et al [41] presented a study on long running processes in the context of melanoma surveillance; Fernandez-Llatas et al [42] presented a study analyzing medical emergency processes; Badakhshan [43] et al presented a study of PM for process analysis, conformance and improvement for the process of pre-hospital emergency department; Tamburis et al [44] presented an approach to investigate conformance between a log file and a simulation tool's generated data while linking PM to discrete event simulation modeling; Kukreja et al [45] applied PM4CA to compare different PM approaches using a sepsis case study; Van Dongen et al [46] presented a conformance checking approach focused on mixed-paradigm process models; Asare et al [47] presented a conformance analysis between processes and the workflows on hospitals; Helm et al [48] introduced a modeling representation method based on multi-perspective declarative PM and a novel algorithm to trace and verify conformance; and Marazza et al [49] presented an approach to compare process models for patient populations and a case study in breast cancer care, by using cross-log conformance checking and standard graph similarity metrics.…”