Process Mining in Healthcare Ronny Mans Introduction This talk: - PowerPoint PPT Presentation

Process Mining in Healthcare Ronny Mans

Introduction This talk: • Applicability of Process Mining in the healthcare domain • Challenges -> Results from applying Process Mining in the AMC 23-9-2010 PAGE 1

Overview • Introduction • AMC • AMC case study • DBCs gynecological oncology • DBCs GO + radiotherapy + chemotherapy • Future work • Conclusion • Questions / Discussion 23-9-2010 PAGE 2

Academic Medical Center (AMC) • University hospital, Amsterdam • 1002 beds • 25.000 patients admitted • 35.000 day admissions • 350.000 outpatient clinic visits • Patients • Own region • Outside region 23-9-2010 PAGE 3

AMC AND • Healthcare processes are highly variable -> Not known what happens in a healthcare process • Process improvement projects • Time consuming to collect data 23-9-2010 PAGE 4

Introduction Data? • Each department has their system • Integration often difficult • Payment system contains all events • DBC: Diagnosis Treatment Combination • Episode of care • All the care steps (inpatient, outpatient, day, and after care) that may be delivered for a specific medical problem or condition. • Each service delivered to a patient is linked to one DBC 23-9-2010 PAGE 5

Data? DBC code: Ovarian cancer DBC code: Cervical cancer Patient: Rose Patient: Sue • Visit outpatient clinic • Visit outpatient clinic • Pathology • Lab test • Lab test • X-ray • MRI • CT • MRI 23-9-2010 PAGE 6

Overview • Introduction • AMC • AMC case study • DBCs gynecological oncology • DBCs GO + radiotherapy + chemotherapy • Conclusion • Future work • Questions / Discussion 23-9-2010 PAGE 7

Case Case – AMC log • 627 Gynaecological oncology patients • 376 Events • 24331 Audit trail entries • Gynaecology, Nursing wards, Radiology, … • All care steps for GO patients 23-9-2010 PAGE 8

Data Only day timestamps 23-9-2010 PAGE 9

Data Single lab tests 23-9-2010 PAGE 10

Data Visit to one department 23-9-2010 PAGE 11

Log Preprocessing Filtering: Getting the right abstraction • Remapping • Aggregation 23-9-2010 PAGE 12

Filtering • Representative R • Keep R • Remove others Remap Element Log 23-9-2010 PAGE 13

Example Remap Element Log Acceptance Na Ka Xray Hg Lab Test Acceptance Xray Hg Na O2-sat Lab Test 23-9-2010 PAGE 14

Filtering • No Representative R • Define R • Remap all to R Remap Element Log • Aggregate lab representative repetitions in trace lab representative lab representative Repetition to lab representative Activity lab representative lab representative lab representative 23-9-2010 PAGE 15

Example Remap Element Log Repetition to Activity lab R Na lab R Ka Xray lab R Hg Xray lab R lab R Hg lab R Na O2-sat 23-9-2010 PAGE 16

Case Result after filtering 23-9-2010 PAGE 17

Case Log Preprocessing (2) Clustering: Grouping similar behaviour • Trace clustering plug-in 23-9-2010 PAGE 18

Case Case – clusters 23-9-2010 PAGE 19

Case Case – biggest cluster 23-9-2010 PAGE 20

Case Case – Social Network Interaction with dietics department 23-9-2010 PAGE 21

Case Case – Basic Performance Analysis Plug-in 23-9-2010 PAGE 22

Results Results so far • Complex hospital logs can be mined • Log pre-processing can be used to derive understandable models • Filtering – for getting the right abstraction • Clustering – for analysing common behaviour 23-9-2010 PAGE 23

Overview • Motivation • AMC case study • DBCs gynecological oncology • DBCs GO + radiotherapy + chemotherapy • Future work • Conclusion • Questions 23-9-2010 PAGE 24

AMC data • DBC: Diagnosis Treatment Combination • Episode of care • All the care steps (inpatient, outpatient, day, and after care) that may be delivered for a specific medical problem or condition. • Each service delivered to a patient is linked to one DBC • Whole care path of gynecological oncology • Gynecological oncology • Radiotherapy • Internal medicine 23-9-2010 PAGE 25

DBCs DBC code: GO Ovarian cancer DBC code: GO Cervical cancer Patient: Rose Patient: Sue • Visit outpatient clinic • Visit outpatient clinic • Pathology • Lab test • Lab test • X-ray • MRI • CT DBC code: gynaecological tumors DBC code: IM Cervical cancer • MRI Patient: Sue Patient: Sue • Visit outpatient clinic • Visit outpatient clinic • Radiotherapy • Lab test • MRI • Chemo therapy • Radiotherapy • Lab test • MRI • Chemo therapy 23-9-2010 PAGE 26

AMC data • Resulting log: • 682 Gynaecological oncology patients • 43615 Audit trail entries • Gynaecology, Nursing wards, Radiology, Radiotherapy, … 23-9-2010 PAGE 27

Diagnostic + therapeutic process • Visual insights • Process Mining 23-9-2010 PAGE 28

Visual insights Radiology Lab Visit to the outpatient clinic 23-9-2010 PAGE 29

Visual insights 23-9-2010 PAGE 30

First surgery Visual insights 23-9-2010 PAGE 31 31

Process Mining • Split log in two parts • Diagnostic part • Therapeutic part 23-9-2010 PAGE 32

Diagnostic process 23-9-2010 PAGE 33

ArtificialStartTask (complete) 100 Diagnostic process (HM) 0,944 0,974 32 54 Pathology 0,978 (complete) 60 46 OC Gyn Onc 0,947 (complete) 0,944 148 43 47 • Focus on most Lab (complete) 0,983 215 important events 69 0,987 76 Pharmacy Lab (complete) 0,917 76 95 0,982 67 Radiology (complete) 0,947 161 64 • Fitness: 0,7 0,809 80 Nursing Ward H5Z (complete) 0,984 • Performance 263 167 0,989 related data 96 Operating Rooms (complete) 100 0,99 100 ArtificialEndTask (complete) 23-9-2010 PAGE 34 100

Overview • Introduction • AMC • AMC case study • Future work • Conclusion • Questions / Discussion 23-9-2010 PAGE 35

Research Goal: Obtain understandable results for the process analyst and medical specialist (end-user) 1. Capturing flexible processes 2. Presentation of process related information 23-9-2010 PAGE 36

Flexible processes (Declarative PLs) 23-9-2010 PAGE 37

Flexible processes (Declarative PLs) 23-9-2010 PAGE 38

Flexible processes Develop mining techniques for less procedural and declarative languages 23-9-2010 PAGE 39

Process related information 23-9-2010 PAGE 40

Process related information • Good maps? • Navigation by PowerPoints? • Traffic information? • Where is the next fuel station? • Who is in charge? • Seamless zoom? • Customizable views? • When will the destination be reached? 23-9-2010 PAGE 41

Overview • Introduction • AMC • AMC case study • Future work • Conclusion • Questions / Discussion 23-9-2010 PAGE 42

Results Conclusions • Insights into healthcare processes • Remaining challenges Capturing flexible processes • • Presentation of process related information • Only day timestamps 23-9-2010 PAGE 43

Questions? / Discussion 23-9-2010 PAGE 44