Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Analyzing the Trajectories of Patients with Sepsis using Process Mining


Published on

Presented at BPMDS 2017:

F. Mannhardt, D. Blinde (2017). Analyzing the Trajectories of Patients with Sepsis using Process Mining. In RADAR+EMISA 2017, pp. 72–80.

Process mining techniques analyze processes based on event data. We analyzed the trajectories of patients in a Dutch hospital from their registration in the emergency room until their discharge. We considered a sample of 1050 patients with symptoms of a sepsis condition, which is a life-threatening condition. We extracted an event log that includes events on activities in the emergency room, admission to hospital wards, and discharge. The event log was enriched with data from laboratory tests and triage checklists. We tried to automatically discover a process model of the patient trajectories, we check conformance to medical guidelines for sepsis patients, and visualize the flow of patients on a de-jure process model. The lessons-learned from this analysis are: (1) process mining can be used to clarify the patient flow in a hospital; (2) process mining can be used to check the daily clinical practice against medical guidelines; (3) process discovery methods may return unsuitable models that are difficult to understand for stakeholders; and (4) process mining is an iterative process, e.g., data quality issues are often discovered and need to be addressed.

Published in: Science
  • Be the first to comment

  • Be the first to like this

Analyzing the Trajectories of Patients with Sepsis using Process Mining

  1. 1. Analyzing the Trajectories of Patients with Sepsis using Process Mining Felix Mannhardt (TU/e), Daan Blinde (Blijnder)
  2. 2. The Project: Sepsis Patients in a Regional Hospital PAGE 1 Sepsis is an inflammation caused by an infection. • Causes poor organ function, insufficient blood flow, high blood lactate • Criteria at least two SIRS criteria due to a presumed infection:  Temp >38°C (100.4°F) or < 36°C (96.8°F)  Heart rate > 90 bpm  Respiratory rate > 20 breaths/min  White blood cell count less than 4000 cells/mm³ • Signs fever, increase heart rate, increase breathing rate, and confusion
  3. 3. The Scope: Sepsis Patient Trajectories PAGE 2
  4. 4. The Event Log PAGE 3
  5. 5. Process Discovery – Inductive Miner PAGE 4 Unsuitable for Communication and Questions • Incomprehensible for stakeholders: Why an exclusive choice: IV Antibiotics and ER Sepsis Triage? • Difficult to answer questions based on that model!
  6. 6. Our Approach • Iteratively design a suitable process model • Visualize patient flow with alignment-based conformance checking techniques • Enhance the model to answer questions: • Q1 – Conformance to medical guideline − Antibiotics within 1 hour − Lactic acid within 3 hours • Q2 – Visualize Patient Trajectories − Admission NC  Admission IC • Q3 – Analyze Returning Patients − Return ER within 28 days PAGE 5
  7. 7. Discovered Data Quality Issue PAGE 6 IV Antibiotics time stamps entered manually Typical mistake found several times 14/06/2017 00:05:00  14/06/2017 23:55:00 13/06/2017 23:55:00
  8. 8. Conclusion & Future work • Scope & Questions • Emergency room of a regional hospital • Trajectories of SEPSIS patient: Conformance, Visualization, and Discovery • Approach • Iteratively designed normative process model • Conformance checking with MPE • Lessons learned • Scoping helps to clarify and visualize patient flow • Medical guidelines can be encoded and checked • Process discovery returned an unsuitable model • Data quality issues discovered by process mining
  9. 9. Questions? PAGE 8 @fmannhardt – –