Explainable Artificial Intelligence and Process Mining Applications for Healthcare (Communications in Computer and Information Science)

✍ Scribed by Jose M. Juarez (editor), Carlos Fernandez-Llatas (editor), Concha Bielza (editor), Owen Johnson (editor), Primoz Kocbek (editor), Pedro Larrañaga (editor), Niels Martin (editor), Jorge Munoz-Gama (editor), Gregor Štiglic (editor)

• Publisher: Springer
• Year: 2024
• Tongue: English
• Leaves: 140
• Category: Library

✦ Synopsis

This book constitutes the proceedings of the Third International Workshop on Explainable Artificial Intelligence in Healthcare, XAI-Healthcare 2023, and the First International Workshop on Process Mining Applications for Healthcare, PM4H 2023, which took place in conjunction with AIME 2023 in Portoroz, Slovenia, on June 15, 2023.
The 7 full papers included from XAI-Healthcare were carefully reviewed and selected from 11 submissions. They focus on all aspects of eXplainable Artificial Intelligence (XAI) in the medical and healthcare field. For PM4H 5 papers have been accepted from 17 submissions. They deal with data-driven process analysis techniques in healthcare.

✦ Table of Contents

Preface
Organization
Contents
International Workshop on Explainable Artificial Intelligence in Healthcare
Unlocking the Power of Explainability in Ranking Systems: A Visual Analytics Approach with XAI Techniques
1 Introduction
2 Related Work
3 Methodology
3.1 XAI
3.2 Interactive Visualization
4 Case Study: Explaining Triaging Patients to Be Admitted to ICU
5 Discussion
6 Conclusion
References
Explainable Artificial Intelligence in Response to the Failures of Musculoskeletal Disorder Rehabilitation
1 Introduction
2 Background and Context of This Work
3 Complexity of the Generation of Self-recovery Exercises
3.1 Defining the Rules
3.2 Application of the Consensus Rules
4 General Structure of Recov’Up
5 Explicability
6 Data Extension and Perspectives
References
An Explainable AI Framework for Treatment Failure Model for Oncology Patients
1 Introduction
2 Scope of Work
2.1 Approaches to Explainability
3 Methodology
3.1 Treatment Failure Explanations
3.2 Model Level Explanations
3.3 Challenges and Limitations
4 Results and Discussion
4.1 Dataset Insights
4.2 Treatment Failure Explanations
4.3 Counterfactual
4.4 Model Level Explanations
5 Conclusion and Future Work
References
Feature Selection in Bipolar Disorder Episode Classification Using Cost-Constrained Methods
1 Introduction
2 Methodology
2.1 Data Preprocessing
2.2 Algorithm
3 Preliminary Results
4 Conclusions and Future Plans
References
ProbExplainer: A Library for Unified Explainability of Probabilistic Models and an Application in Interneuron Classification
1 Introduction
2 Background
2.1 Probabilistic Models: Bayesian Networks
2.2 Existing Software
2.3 Interneuron Classification: The Gardener Approach
3 Software Framework
3.1 A Unified Interface
3.2 Design of the Algorithms
4 Application in GABAergic Interneuron Classification
4.1 Data
4.2 Experiments
4.3 Results
5 Conclusions and Future Work
References
Interpreting Machine Learning Models for Survival Analysis: A Study of Cutaneous Melanoma Using the SEER Database
1 Introduction
2 Surveillance, Epidemiology, and End Results Database
2.1 Selection of the Individuals
2.2 Exploratory Data Analysis
3 Machine Learning Models
3.1 Data Preprocessing
3.2 Machine Learning Models
4 Explainability
5 Conclusions
References
Explanations of Symbolic Reasoning to Effect Patient Persuasion and Education
1 Introduction
2 Derivation Proofs
3 Human-Readable Explanation Generation from Derivation Proofs
3.1 Pre-processing Module
3.2 Describe Module
3.3 Collect Module
4 Demonstration
5 Planned Evaluation
6 Conclusions and Future Work
References
International Workshop on Process Mining Applications for Healthcare
PMApp: An Interactive Process Mining Toolkit for Building Healthcare Dashboards
1 Introduction
2 Through an Interactive Process Mining Solution for Healthcare
3 PMApp: An Interactive Process Mining Toolkit
3.1 Experiment Designer
3.2 Ingestor Editor
3.3 Dashboard
4 Discussion and Conclusions
References
A Data-Driven Framework for Improving Clinical Managements of Severe Paralytic Ileus in ICU: From Path Discovery, Model Generation to Validation
1 Introduction
2 Method
2.1 Data Resource
2.2 Cohort Extraction
2.3 Event Log Extraction
2.4 Frequent Patient Pathways Discovery
2.5 Structural Equation Modelling
3 Results
4 Discussion and Conclusion
References
Phenotypes vs Processes: Understanding the Progression of Complications in Type 2 Diabetes. A Case Study
1 Introduction
2 Methodology
2.1 Data Collection and Study Measures
2.2 Methods
2.3 Data Corpus and Event Log Generation
3 Results
3.1 Patient-Level Analysis
3.2 Short-Term Pathways Analysis
4 Discussion
References
From Script to Application. A bupaR Integration into PMApp for Interactive Process Mining Research
1 Introduction
2 Background
3 Use Case and Test Scenario
3.1 Filtering Cases Based on Activities and Timestamps
3.2 Data Augmentation and Conditional Filtering
3.3 Manipulation of Activities
4 Discussion and Future Work
References
Understanding Prostate Cancer Care Process Using Process Mining: A Case Study
1 Introduction
2 Materials and Methods
3 Results
4 Discussion and Conclusions
References
Author Index