Automatic Process Comparison for Subpopulations: Application in Cancer Care

Abstract

Processes in organisations, such as hospitals, may deviate from the intended standard processes, due to unforeseeable events and the complexity of the organisation. For hospitals, the knowledge of actual patient streams for patient populations (e.g., severe or non-severe cases) is important for quality control and improvement. Process discovery from event data in electronic health records can shed light on the patient flows, but their comparison for different populations is cumbersome and time-consuming. In this paper, we present an approach for the automatic comparison of process models that were extracted from events in electronic health records. Concretely, we propose comparing processes for different patient populations by cross-log conformance checking, and standard graph similarity measures obtained from the directed graph underlying the process model. We perform a user study with 20 participants in order to obtain a ground truth for similarity of process models. We evaluate our approach on two data sets, the publicly available MIMIC database with the focus on different cancer patients in intensive care, and a database on breast cancer patients from a Dutch hospital. In our experiments, we found average fitness to be a good indicator for visual similarity in the ZGT use case, while the average precision and graph edit distance are strongly correlated with visual impression for cancer process models on MIMIC. These results are a call for further research and evaluation for determining which similarity or combination of similarities is needed in which type of process model comparison.

Towards a trustworthy, secure and reliable enclave for machine learning in a hospital setting: The Essen Medical Computing Platform (EMCP)

“…A patient's journey through the hospital might deviate from the guidelines for multiple reasons, such as unforeseeable events and the complexity of the organization. Electronic Health Records (EHRs) contain information about events related to diagnostics and treatment of patients and can serve as input for retracing a patient's journey [4]. The application of tools to compute a patient's journey from observational data is called process mining.…”

“…Conversely, users in mode ii) must not be able to access the NFS file service, as they may assume the role of any user through local root privileges. 4 This way it would be possible 3 docs.docker.com/engine/security/userns-remap/ 4 The authors are aware that using Kerberized NFS would solve this problem but also increase complexity. The available S3-compatible object storage is the primary storage for users with independent platforms.…”

Modelagem, Mineração e Análise de Jornadas / Trajetórias de Pacientes

“…com operações como junções e divisões AND/OR/XOR) são adequados. Nessa linha, são exemplos de modelos aplicados na literatura associada a jornadas de pacientes o BPMN [Kurniati et al 2019, Lu et al 2016, Stefanini et al 2020, Tamburis and Esposito 2020, HeuristicNet [Caron et al 2014, Durojaiye et al 2018, Inductive Visual Model [Andrews et al 2020, Kempa-Liehr et al 2020, Marazza et al 2020], as redes de Petri [Durojaiye et al 2018, Marazza et al 2020, Rebuge and Ferreira 2012, Stefanini et al 2020, Tamburis and Esposito 2020, Fork/Join Network [Senderovich et al 2016] e modelos declarativos de processo [Mertens et al 2018,Mertens et al 2018. Um exemplo desse tipo de modelo, usando a notação BPMN, é apresentado na Figura 3.…”

“…Os modelos de jornada de pacientes não apenas proporcionam uma melhor compreensão de um grupo específico de pacientes, mas permitem compará-los com outros grupos. Marazza et al 2020 propuseram um método para comparar jornadas de duas subpopulações de pacientes com câncer que consiste em obter um modelo de processo para cada grupo, converter os modelos de processo em gráficos direcionados e compará-los com Distância de Edição de Grafos (vide Subseção 3.3.2), ou com um vetor de atributos do grafo. Os autores utilizaram o Inductive Miner para obter jornadas representadas como redes de Petri, e também uma extensão do algoritmo -o Inductive Miner Infrequent -para minerar jornadas representadas com o Inductive Visual Model.…”

Process mining applications in the healthcare domain: A comprehensive review

“…As we will show in the next sections, HIS, EHR, and EMR of healthcare facilities are not the only source of data for PM applications. In some work the event log has been synthetically generated using (Burattin, 2016; Burattin & Alessandro Sperduti, 2010), extracted from video recordings (e.g., of a surgical operation; A. Grando et al, 2017; Kelleher et al, 2014; Lira et al, 2019; S. Yang, Tao, et al, 2018), from body‐sensor data (e.g., sequence of blood pressure measurements; Fernández‐Llatas, Garcia‐Gomez, et al, 2011; Kaymak et al, 2012; McGregor et al, 2011), from location systems such as Real Time Location Systems (RTLS) logs, and from the Medical Information Mart for Intensive Care III (MIMIC‐III) 5 open access dataset (Alharbi et al, 2018; A. P. Kurniati et al, 2018a, 2019; Marazza et al, 2020; Pika et al, 2019; Rojas & Capurro, 2018).…”