healthprocessai

Process Mining in Healthcare: A Clinician’s Guide

Table of Contents

  1. Introduction: What is Process Mining?
  2. Why Process Mining Matters for Healthcare
  3. Understanding the Fundamentals
  4. Process Mining in Emergency Care
  5. Applications in Chronic Disease Management
  6. Optimizing Clinical Pathways
  7. Cancer Care and Treatment Optimization
  8. Surgical and Perioperative Care
  9. Quality Improvement and Patient Safety
  10. Implementation Challenges and Solutions
  11. Future Directions
  12. Getting Started: A Practical Guide

1. Introduction: What is Process Mining?

Process mining is a data-driven approach that automatically discovers how healthcare processes actually work, rather than how we think they work¹. Unlike traditional methods that rely on interviews or observations, process mining analyzes digital footprints left in electronic health records (EHRs), hospital information systems, and other clinical databases to reveal real patient journeys through care.

The Core Concept

Imagine being able to trace every step of every patient’s journey through your hospital — from admission to discharge — and then automatically creating a visual map showing all the different pathways patients take, where delays occur, and which treatments are most effective. This is exactly what process mining does².

Three Types of Process Mining

Type What It Does Clinical Example Key Benefits
Process Discovery Creates process maps from your data Shows actual sepsis care pathways vs. protocols Reveals hidden variations and bottlenecks
Conformance Checking Compares reality vs. guidelines Measures adherence to stroke care protocols Identifies deviation patterns and compliance gaps
Process Enhancement Predicts and optimizes processes Forecasts patient deterioration risk Enables proactive interventions and resource planning
Process Mining Types — Visual Overview:

Historical Data  →  [Process Discovery]  →  "What actually happens?"
    ↓
Guidelines + Reality  →  [Conformance Checking]  →  "Are we following protocols?"
    ↓
Current Process + Predictions  →  [Process Enhancement]  →  "How can we improve?"

2. Why Process Mining Matters for Healthcare

The Healthcare Data Explosion

Modern healthcare generates enormous amounts of data — every medication administration, diagnostic test, consultation, and patient transfer is recorded digitally³. However, this wealth of information often remains trapped in separate systems, making it difficult to understand the complete patient experience.

Traditional Process Analysis Limitations

Conventional approaches to understanding healthcare processes have significant limitations:

What Process Mining Reveals

Process mining transforms your existing clinical data into actionable insights by revealing:

Insight Category What You Discover Clinical Impact
Hidden Inefficiencies Bottlenecks causing delays; Unnecessary duplicate steps; Resource constraints Reduced patient wait times; Lower operational costs; Improved staff satisfaction
Practice Variations Different approaches to similar cases; Pathway-outcome relationships; Guideline adherence patterns Standardization opportunities; Best practice identification; Quality improvement targets
Real Patient Journeys Complete care experiences; Department transitions; Time in different care states Patient-centered improvements; Coordination optimization; Experience enhancement
Traditional Analysis vs. Process Mining:

Traditional Approach:    Process Mining Approach:
┌─────────────────┐    ┌─────────────────┐
│ Manual Surveys    │    │ Automated Data    │
│ Small Samples    │    vs.    │ Complete Records│
│ Perceptions    │    │ Actual Events    │
│ Static View    │    │ Dynamic Process │
└─────────────────┘    └─────────────────┘
    ↓    ↓
    Limited Insight    Comprehensive Understanding

3. Understanding the Fundamentals

Event Logs: The Foundation

Process mining works by analyzing “event logs” — timestamped records of activities that occur during patient care. Every electronic health record system automatically generates these logs, capturing:

Element Example
Patient ID Patient_12345
Activity Blood culture drawn
Timestamp 2024-03-15 14:30:00
Resource Lab technician_A
Location Emergency department

From Data to Insights

The process mining analysis transforms these individual events into comprehensive process maps that show:

Event Log Transformation Process:

Raw Events:    Process Map:    Clinical Insights:
┌──────────────┐    ┌─────────────────┐    ┌──────────────────┐
│Patient_001    │    │    [Triage]    │    │• Average wait time│
│Triage    │    │    ↓    │    │• Bottleneck points│
│08:00 AM    │    ──── →     │    [Blood Work]    │    ──── →     │• Best pathways    │
│    │    │    ↓    │    │• Outcome patterns │
│Patient_001    │    │    [Diagnosis]    │    │• Resource needs    │
│Blood Work    │    │    ↓    │    │    │
│08:45 AM    │    │    [Treatment]    │    │    │
│...    │    └─────────────────┘    └──────────────────┘
└──────────────┘
Process Mining Output Clinical Application Decision Support
Process Flow Diagrams Visual patient pathways Identify optimal care routes
Performance Metrics Time and resource analysis Allocate staff efficiently
Variant Analysis Common vs. rare pathways Standardize frequent cases
Bottleneck Detection Delay identification Target improvement efforts

4. Process Mining in Emergency Care

Emergency Department Optimization

Emergency departments are ideal settings for process mining due to their complex, time-sensitive processes and rich digital footprints⁴.

ED Process Mining Applications Key Findings Impact Metrics
Patient Flow Analysis Triage bottlenecks identified 15-25% reduction in wait times
Laboratory Integration Blood work delays discovered 30% faster turnaround times
Discharge Processes Complex procedures streamlined 20% reduction in length of stay
Resource Allocation Staffing patterns optimized 18% improvement in efficiency
Typical ED Patient Flow — Before vs. After Process Mining:

BEFORE (Traditional View):
Arrival  →  Triage  →  Waiting  →  Doctor  →  Tests  →  Results  →  Discharge
    ↓    ↓    ↓    ↓    ↓    ↓    ↓
 5 min    15 min    45 min    20 min    30 min    60 min    15 min
    Total Time: 3h 10min

AFTER (Process Mining Optimized):
Arrival  →  Fast-Track Triage  →  Parallel Processing  →  Discharge
    ↓    ↓    ↓    ↓
 5 min    8 min    ┌─ Doctor (15 min)
    ├─ Tests (20 min)     →  10 min
    └─ Results (25 min)
    Total Time: 1h 23min (56% improvement)

Case Study: Sepsis Care A landmark study of 1,050 sepsis cases revealed that only 66-77% of patients followed the intended clinical pathway during their first cycle of treatment⁵.

Sepsis Pathway Analysis Finding Clinical Significance
Adherence Rate 66-77% follow protocol 23-34% variation needs investigation
Critical Time Points 6-12 hour intervention window Early detection saves lives
Resource Correlation Staffing patterns predict outcomes Evidence for optimal staffing
Mortality Factors Workflow variations increase risk Process standardization critical

Stroke Care Pathways

Process mining applications in stroke care have demonstrated particular value in time-critical situations⁶. Analysis of stroke patient pathways revealed:

Time-to-Treatment Optimization

Quality Improvement


5. Applications in Chronic Disease Management

Longitudinal Patient Monitoring

Process mining excels in analyzing chronic disease progression because it can track patients over extended periods⁷. The methodology bridges traditional epidemiological approaches with data-driven process analysis.

Chronic Disease Process Mining Framework:

Traditional Epidemiology:    Process Mining Enhancement:
┌─────────────────────┐    ┌─────────────────────────────┐
│ Exposure  →  Outcome    │    │ Drug  →  Decline  →  KRT  →  Death │
│    │    +    │    ↓    ↓    ↓    │
│ Single Events    │    │    Time Analysis    │    Risk    │
│ Statistical Tests    │    │    Pathways    │    Factors    │
└─────────────────────┘    │    Predictions    │    │
    └─────────────────────────────┘

Chronic Kidney Disease Example A recent study applied process mining to analyze kidney function progression in patients taking proton pump inhibitors (PPIs) versus H2 blockers⁸:

Study Component PPI Group H2B Group Risk Difference
Patient Count 100,803 9,774
30% eGFR Decline 9.02% 3.37% +168%
All-cause Mortality 12.06% 2.16% +458%
Kidney Replacement Therapy 0.16% 0% N/A
Risk Analysis Hazard Ratio (95% CI) Clinical Interpretation
Kidney Function Decline 1.6 (1.4-1.8) 60% increased risk with PPIs
Mortality Risk 3.0 (2.1-4.4) 3-fold higher death rate
Combined Endpoint 1.8 (1.6-2.1) Significant overall harm
Medication Pathway Comparison:

PPI Pathway:    H2B Pathway:
Drug Initiation    Drug Initiation
    ↓    ↓
Higher Inflammation    Lower Inflammation    
    ↓    ↓
Kidney Damage (9.02%)    Kidney Stability (3.37%)
    ↓    ↓
Higher Mortality (12.06%)    Lower Mortality (2.16%)

Key Insight: Process mining revealed that medication choice
creates distinct disease progression pathways

Diabetes Care Pathways

Process mining in diabetes management has revealed:

Cardiovascular Disease Management

Applications include:


6. Optimizing Clinical Pathways

From Idealized to Realistic Pathways

Traditional clinical pathways are often designed based on idealized patient scenarios. Process mining reveals the reality of clinical practice⁹:

Pathway Adherence Reality Percentage Clinical Implication
Exact Protocol Adherence 5-30% Most patients require individualization
Core Elements Adherence 60-80% High-performing pathways maintain essentials
Clinically Appropriate Deviations 40-60% Proper individualized care
Process Inefficiencies 10-25% Improvement opportunities
Idealized vs. Reality — Breast Cancer Treatment:

IDEALIZED PATHWAY:
Diagnosis  →  Chemo Cycle 1  →  Cycle 2  →  ...  →  Cycle 6  →  Recovery
    ↓    ↓    ↓    ↓    ↓
Expected    Expected    Expected    Expected    Planned
Timeline    Timeline    Timeline    Timeline    Outcome

ACTUAL PATHWAYS (474 variants discovered):

Patient Type A (5%): Diagnosis  →  C1  →  C2  →  C3  →  C4  →  C5  →  C6  →  Recovery
Patient Type B (15%): Diagnosis  →  C1  →  [Hospital]  →  C2  →  C3  →  [Delay]  →  C4...
Patient Type C (25%): Diagnosis  →  C1  →  C2  →  [Toxicity]  →  Dose Adjustment  →  C3...
Patient Type D (31%): Diagnosis  →  C1  →  [Emergency]  →  Recovery  →  C2...
...and 470 more variants

KEY INSIGHT: Only 5% followed the "standard" 6-cycle pathway without complications

Breast Cancer Treatment Analysis

A comprehensive study of breast cancer patients revealed significant insights¹⁰:

Treatment Reality Breast Cancer (535 patients) Colorectal Cancer (420 patients)
Pathway Variants 474 different pathways 329 different pathways
Perfect Adherence 27 patients (5%) 26 patients (6%)
Hospital Admissions 169 patients (31.6%) 190 patients (45.2%)
Unplanned Contacts 95% had deviations 94% had deviations
Cancer Treatment Process Map (Simplified):

Standard Expected:
[Diagnosis]  →  [Chemo 1]  →  [Chemo 2]  →  [Chemo 3]  →  [Chemo 4]  →  [Chemo 5]  →  [Chemo 6]

Reality Discovered:
    ┌ →  [Emergency Admission]  →  [Recovery] ──┐
[Diagnosis]  →  [Chemo 1] ┼ →  [Dose Reduction]  →  [Chemo 2] ──────┼ →  [Chemo 3]
    └ →  [Toxicity Management]  →  [Delay] ────────┘
    ↓
    ┌ →  [Hospital Stay]  →  [Chemo 4] ─────────┐
    [Chemo 3] ──┼ →  [Outpatient]  →  [Chemo 4] ────────────┼ →  [Assessment]
    └ →  [Treatment Break]  →  [Restart] ────────┘

Process mining revealed: The "simple" pathway is actually highly complex
with multiple decision points and recovery loops.

Surgical Pathway Optimization

Process mining applications in surgery include:


7. Cancer Care and Treatment Optimization

Chemotherapy Pathway Analysis

Process mining has proven particularly valuable in oncology due to the complexity and variability of cancer treatment pathways¹¹:

Treatment Compliance Monitoring

Toxicity Management

Melanoma Surveillance

Long-term surveillance processes for cancer survivors benefit from process mining analysis¹²:


8. Surgical and Perioperative Care

Operating Room Efficiency

Process mining applications in surgical settings focus on:

Case Scheduling Optimization

Perioperative Workflow

Complication Prevention

Process mining helps identify:


9. Quality Improvement and Patient Safety

Medication Safety

Process mining applications in medication management include¹³:

Prescription to Administration Analysis

Adverse Drug Event Detection

Infection Control

Healthcare-associated infection prevention benefits from process mining through:


10. Implementation Challenges and Solutions

Common Challenges

Challenge Category Specific Issues Impact Level Frequency
Data Quality Incomplete documentation; Missing timestamps; Inconsistent coding; Multi-system data High 80% of projects
Technical Complexity Specialized software needs; IT integration requirements; Staff training demands Medium 60% of projects
Organizational Resistance Workflow concerns; Monitoring anxiety; Unclear value proposition High 70% of projects
Resource Constraints Time limitations; Budget restrictions; Competing priorities Medium 90% of projects
Implementation Challenge Framework:

Technical Challenges    Organizational Challenges    Clinical Challenges
┌─────────────────┐    ┌─────────────────────────┐    ┌─────────────────┐
│• Data Quality    │    │• Change Resistance    │    │• Clinical Buy-in│
│• System Integration│     →     │• Resource Allocation    │     →     │• Interpretation │
│• Software Tools │    │• Leadership Support    │    │• Action Planning│
│• Training Needs │    │• Communication Issues    │    │• Sustainability │
└─────────────────┘    └─────────────────────────┘    └─────────────────┘
    ↓    ↓    ↓
    Address First    Manage Continuously    Focus on Value

Solutions and Best Practices

Challenge Solution Strategy Implementation Steps Success Factors
Data Quality Iterative assessment and improvement 1. Audit current data; 2. Identify gaps; 3. Implement fixes; 4. Validate improvements Strong IT partnership
Technical Skills Build internal capability 1. Train core team; 2. Start with simple tools; 3. Gradual complexity; 4. External support initially Dedicated resources
Resistance Clinical leadership engagement 1. Identify champions; 2. Show early wins; 3. Address concerns; 4. Celebrate successes Transparent communication
Sustainability Embedded processes 1. Regular analysis cycles; 2. Automated reporting; 3. Continuous improvement; 4. Knowledge transfer Executive support
Successful Implementation Roadmap:

Phase 1: Foundation (Months 1-2)
┌─────────────────────────────────┐
│ ✓ Assess data quality    │
│ ✓ Form core team    │
│ ✓ Select initial process    │
│ ✓ Secure leadership support    │
└─────────────────────────────────┘
    ↓
Phase 2: Pilot (Months 3-4)
┌─────────────────────────────────┐
│ ✓ Extract and analyze data    │
│ ✓ Generate initial insights    │
│ ✓ Validate with clinicians    │
│ ✓ Identify improvement opportunities│
└─────────────────────────────────┘
    ↓
Phase 3: Implementation (Months 5-8)
┌─────────────────────────────────┐
│ ✓ Pilot process changes    │
│ ✓ Measure impact    │
│ ✓ Refine based on feedback    │
│ ✓ Scale successful changes    │
└─────────────────────────────────┘
    ↓
Phase 4: Sustainability (Ongoing)
┌─────────────────────────────────┐
│ ✓ Regular monitoring cycles    │
│ ✓ Continuous improvement    │
│ ✓ Expand to other processes    │
│ ✓ Build organizational capability│
└─────────────────────────────────┘

11. Future Directions

Artificial Intelligence Integration

The combination of process mining with AI and machine learning opens new possibilities¹⁴:

Predictive Process Monitoring

Personalized Medicine Integration

Population Health Applications

Process mining is expanding beyond individual hospitals to:

Continuous Learning Systems

Future developments include:


12. Getting Started: A Practical Guide

Step 1: Identify the Right Process

Process Characteristics Good Candidates ✅ Poor Candidates ❌
Volume High-volume (>100 cases/month) Low-volume (<20 cases/month)
Standardization Standardized protocols exist Highly individualized care
Digital Documentation Well-documented electronically Primarily paper-based
Clinical Leadership Champion identified No clinical interest
Improvement Potential Known quality/efficiency issues Already optimized processes
Complexity Moderate complexity Extremely simple or complex
Process Selection Matrix:

High Volume + High Digital Documentation = IDEAL
    ↑    ↑
    │    │
    │    GOOD CHOICE    │
    │    ┌──────────┴──────────┐
    │    │    │
    │    │    • Emergency Dept    │
    │    │    • Surgery Scheduling│
    │    │    • Medication Admin │
    │    │    • Discharge Process │
    │    │    │
    │    └─────────────────────┘
    │
Low Volume + Poor Documentation = AVOID

Excellent Starting Points:

Process Type Why It Works Expected Benefits
Emergency Department High volume, time-sensitive, well-documented 15-25% wait time reduction
Surgical Scheduling Standardized protocols, resource-intensive 20-30% efficiency gain
Medication Administration Safety-critical, frequent activities 30-50% error reduction
Discharge Process Cross-departmental, improvement opportunities 10-20% length of stay reduction

Step 2: Assess Data Readiness

Data Element Essential ✓ Nice to Have ○ Assessment Questions
Patient ID   Can you link events to specific patients?
Activity Names   Are activities standardized and meaningful?
Timestamps   Are all events time-stamped accurately?
Outcomes   Can you measure process success?
Resources   Who performed each activity?
Locations   Where did activities occur?
Clinical Data   Lab values, vital signs, diagnoses?
Data Quality Assessment Framework:

Step 1: Data Inventory    Step 2: Quality Check    Step 3: Gap Analysis
┌─────────────────────┐    ┌─────────────────────┐    ┌─────────────────────┐
│ □ Patient IDs    │    │ Completeness: ___%    │    │ Critical Gaps:    │
│ □ Activity codes    │     →     │ Consistency: ___%    │     →     │ □ Missing timestamps│
│ □ Timestamps    │    │ Accuracy: _____%    │    │ □ Inconsistent codes│
│ □ Outcomes    │    │ Linkability: ___%    │    │ □ Data silos    │
│ □ Resources    │    │    │    │ □ Access issues    │
└─────────────────────┘    └─────────────────────┘    └─────────────────────┘
    ↓    ↓    ↓
    What do we have?    How good is it?    What needs fixing?

Step 3: Build the Right Team

Role Responsibilities Time Commitment Key Qualifications
Clinical Champion Process expertise; Staff engagement; Results interpretation 25% for 6 months Respected clinician; Change leadership; Quality improvement experience
Data Analyst Technical analysis; Tool operation; Visualization creation 50% for 6 months Process mining experience; Healthcare data knowledge; Statistical skills
IT Representative Data extraction; System integration; Technical support 15% for 6 months Database expertise; Healthcare systems knowledge; Security awareness
QI Leader Project management; Change facilitation; Implementation planning 30% for 6 months Quality improvement experience; Change management; Healthcare operations
Team Structure and Communication Flow:

    Executive Sponsor
    │
    Project Steering Committee
    │
    ┌──────────────┼──────────────┐
    │    │    │
    Clinical Champion    QI Leader    IT Representative
    │    │    │
    └──────┬───────┼───────┬──────┘
    │    │    │
    Data Analyst │    Frontline Staff
    │
    Patient Representatives

Meeting Cadence:
• Weekly core team meetings (1 hour)
• Bi-weekly stakeholder updates (30 minutes)    
• Monthly steering committee reports (1 hour)

Step 4: Define Success Metrics

Metric Category Examples Measurement Method Target Improvement
Time-Based Cycle time; Wait times; Time to treatment Event log analysis 15-30% reduction
Quality Guideline adherence; Complication rates; Readmissions Clinical indicators 10-25% improvement
Efficiency Resource utilization; Throughput; Cost per case Operational metrics 20-40% optimization
Patient Experience Satisfaction scores; Communication quality; Care coordination Patient surveys 15-25% increase
Measurement Framework — Balanced Scorecard Approach:

Clinical Outcomes    Process Efficiency
┌─────────────────┐    ┌─────────────────┐
│• Mortality    │    │• Cycle time    │
│• Complications    │    │• Throughput    │
│• Readmissions    │    │• Utilization    │
│• Quality scores │    │• Wait times    │
└─────────────────┘    └─────────────────┘
    │    │
    └────────┬─────────────────┘
    │
    ┌─────────────────┐
    │ Patient    │
    │ Experience    │
    │• Satisfaction    │
    │• Communication    │
    │• Care quality    │
    └─────────────────┘
    │
    ┌─────────────────┐
    │ Financial    │
    │ Impact    │
    │• Cost per case    │
    │• Resource costs │
    │• Revenue cycle    │
    └─────────────────┘

Target: 15-25% improvement across all quadrants

Step 5: Conduct Pilot Analysis

Initial Scope:

Early Activities:

Step 6: Interpret Results Clinically

Clinical Validation:

Step 7: Implement Changes

Change Management:

Step 8: Sustain and Scale

Continuous Monitoring:

Scaling Decisions:


Conclusion

Process mining represents a fundamental shift in how we understand and improve healthcare delivery. By leveraging the wealth of data already captured in our clinical information systems, we can move beyond assumptions and anecdotes to evidence-based process improvement.

The Value Proposition for Clinicians

Traditional Approach Process Mining Approach Clinical Advantage
Limited View Complete Picture See entire patient journey
Assumptions Data-Driven Facts Evidence-based decisions
Reactive Predictive Prevent problems before they occur
Anecdotal Systematic Measurable improvement
Siloed Integrated Cross-departmental insights
Process Mining Impact Across Healthcare:

Emergency Care    Chronic Disease    Cancer Treatment
┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│ 22% LOS reduction│    │ 60% risk increase│    │ 474 pathway    │
│ 15-25% wait time │     →     │ identified (PPIs)│     →     │ variants found    │    
│ 18% efficiency    │    │ 3x mortality risk│    │ 95% had    │
│ improvement    │    │ discovered    │    │ deviations    │
└─────────────────┘    └─────────────────┘    └─────────────────┘
    ↓    ↓    ↓
Process optimization    Risk identification    Pathway realism

    Combined Clinical Value:
    ┌─────────────────────────────────┐
    │ • Improved patient outcomes    │
    │ • Reduced costs and waste    │
    │ • Enhanced care coordination    │
    │ • Evidence-based protocols    │
    │ • Predictive capabilities    │
    └─────────────────────────────────┘

Implementation Success Factors

Success Factor Critical Elements Measurement Approach
Clinical Leadership Champion engagement; Frontline involvement; Change readiness Stakeholder surveys; Participation rates; Feedback sessions
Data Quality Complete event logs; Accurate timestamps; Consistent coding Completeness metrics; Accuracy assessments; Quality scores
Technical Capability Skilled analysts; Appropriate tools; IT integration Analysis quality; Tool effectiveness; System integration
Organizational Support Executive backing; Resource allocation; Culture alignment Budget approval; Time allocation; Cultural surveys

Future Opportunities

The technology is mature, the data exists in most healthcare organizations, and the potential for improving patient outcomes is substantial. The question is not whether process mining will transform healthcare, but how quickly we can harness its power to benefit our patients.

Evolution of Healthcare Process Mining:

Current State (2024)    Near Future (2025-2027)    Long-term Vision (2028+)
┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│• Descriptive    │    │• Predictive    │    │• Prescriptive    │
│• Reactive    │     →     │• Real-time    │     →     │• Autonomous    │
│• Single process │    │• Multi-process    │    │• AI-integrated    │
│• Manual analysis│    │• Automated    │    │• Self-learning    │
└─────────────────┘    └─────────────────┘    └─────────────────┘
    ↓    ↓    ↓
"What happened?"    "What will happen?"    "What should happen?"

As we move toward more data-driven healthcare, clinicians who understand and embrace process mining will be better positioned to lead quality improvement efforts, optimize patient care, and demonstrate the value of their clinical expertise. The future of healthcare improvement lies not in replacing clinical judgment with algorithms, but in augmenting clinical expertise with data-driven insights — and process mining is one of the most powerful tools available for this purpose.

Key Takeaways for Clinicians

Takeaway Action Item Expected Outcome
Start Small Choose one high-impact process Quick wins build momentum
Engage Early Partner with analysts and IT Clinically relevant insights
Focus on Value Target patient-centered improvements Meaningful care enhancement
Build Capability Develop internal expertise Sustainable improvement
Share Success Communicate results widely Organizational transformation

The journey begins with a single step — identifying one process that matters to your patients and your practice, and discovering what your data has been trying to tell you all along.


References

  1. Rojas, E., Munoz-Gama, J., Sepúlveda, M., & Capurro, D. (2016). Process mining in healthcare: A literature review. Journal of Biomedical Informatics, 61, 224-236. https://doi.org/10.1016/j.jbi.2016.04.007

  2. De Roock, E., & Martin, N. (2022). Process mining in healthcare — An updated perspective on the state of the art. Journal of Biomedical Informatics, 127, 103995. https://doi.org/10.1016/j.jbi.2022.103995

  3. Aversano, L., Iammarino, M., Madau, A., Pirlo, G., & Semeraro, G. (2025). Process mining applications in healthcare: a systematic literature review. PeerJ Computer Science, 11, e2613. https://doi.org/10.7717/peerj-cs.2613

  4. Williams, R., Rojas, E., Peek, N., & Johnson, O. A. (2018). Process mining in primary care: A literature review. Studies in Health Technology and Informatics, 247, 376-380.

  5. Chen, K., Abtahi, F., Carrero, J. J., Fernandez-Llatas, C., & Seoane, F. (2024). Validation of an interactive process mining methodology for clinical epidemiology through a cohort study on chronic kidney disease progression. Scientific Reports, 14, 27997. https://doi.org/10.1038/s41598-024-79704-5

  6. Mans, R., Schonenberg, H., Leonardi, G., Panzarasa, S., Cavallini, A., Quaglini, S., & van der Aalst, W. M. P. (2008). Process mining techniques: an application to stroke care. Studies in Health Technology and Informatics, 136, 573-578.

  7. Hendricks, G. (2019). Process mining of incoming patients with sepsis. Conference Proceedings — IEEE SOUTHEASTCON, 2019.

  8. Chen, K., Abtahi, F., Carrero, J. J., Fernandez-Llatas, C., & Seoane, F. (2024). Validation of an interactive process mining methodology for clinical epidemiology through a cohort study on chronic kidney disease progression. Scientific Reports, 14, 27997. https://doi.org/10.1038/s41598-024-79704-5

  9. Yang, W., & Su, Q. (2014). Process mining for clinical pathway: Literature review and future directions. 11th International Conference on Service Systems and Service Management (ICSSSM), 1-5.

  10. Baker, K., Dunwoodie, E., Jones, R. G., Newsham, A., Johnson, O., Price, C. P., Wolstenholme, J., Leal, J., McGinley, P., Twelves, C., & Hall, G. (2017). Process mining routinely collected electronic health records to define real-life clinical pathways during chemotherapy. International Journal of Medical Informatics, 103, 32-41. https://doi.org/10.1016/j.ijmedinf.2017.03.011

  11. Wicky, A., Gatta, R., Latifyan, S., Micheli, R., Gerard, C., Pradervand, S., Michielin, O., & Cuendet, M. A. (2023). Interactive process mining of cancer treatment sequences with melanoma real-world data. Frontiers in Oncology, 13, 1043683. https://doi.org/10.3389/fonc.2023.1043683

  12. Rinner, C., Helm, E., Dunkl, R., Kittler, H., & Rinderle-Ma, S. (2018). Process mining and conformance checking of long running processes in the context of melanoma surveillance. International Journal of Environmental Research and Public Health, 15(12), 2809. https://doi.org/10.3390/ijerph15122809

  13. Fernandez-Llatas, C. (Ed.). (2021). Interactive Process Mining in Healthcare. Springer International Publishing.

  14. Tavazzi, E., Daberdaku, S., Vasta, R., Calvo, A., Mora, G., & Rizzo, G. (2023). Leveraging process mining for modeling progression trajectories in amyotrophic lateral sclerosis. BMC Medical Informatics and Decision Making, 22, 346. https://doi.org/10.1186/s12911-023-02113-7