Interpretable machine learning for time-to-event prediction in medicine and healthcare

Time-to-event prediction, e.g. cancer survival analysis or hospital length of stay, is a highly prominent machine learning task in medical and healthcare applications. However, only a few interpretable machine learning methods comply with its challenges. To facilitate a comprehensive explanatory ana...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-03
Main Authors: Baniecki, Hubert, Sobieski, Bartlomiej, Szatkowski, Patryk, Bombinski, Przemyslaw, Biecek, Przemyslaw
Format: Article
Language:English
Subjects:
Algorithms
Cancer
Datasets
Decision making
Feature extraction
Length of stay
Machine learning
Mathematical models
Medical imaging
Modal data
Radiomics
Source code
Survival analysis
Time dependence
Trustworthiness
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Time-to-event prediction, e.g. cancer survival analysis or hospital length of stay, is a highly prominent machine learning task in medical and healthcare applications. However, only a few interpretable machine learning methods comply with its challenges. To facilitate a comprehensive explanatory analysis of survival models, we formally introduce time-dependent feature effects and global feature importance explanations. We show how post-hoc interpretation methods allow for finding biases in AI systems predicting length of stay using a novel multi-modal dataset created from 1235 X-ray images with textual radiology reports annotated by human experts. Moreover, we evaluate cancer survival models beyond predictive performance to include the importance of multi-omics feature groups based on a large-scale benchmark comprising 11 datasets from The Cancer Genome Atlas (TCGA). Model developers can use the proposed methods to debug and improve machine learning algorithms, while physicians can discover disease biomarkers and assess their significance. We hope the contributed open data and code resources facilitate future work in the emerging research direction of explainable survival analysis.
ISSN:2331-8422
DOI:10.48550/arxiv.2303.09817