Open access · CC-BY
via OpenAlex
Interpretable machine learning prediction of all-cause mortality
Wei Qiu, Hugh Chen, Ayse B. Dincer, Scott Lundberg, Matt R. Kaeberlein, Su‐In Lee
Communications Medicine · 2022 · ▲ 111 citations
Abstract
Background: Unlike linear models which are traditionally used to study all-cause mortality, complex machine learning models can capture non-linear interrelations and provide opportunities to identify unexplored risk factors. Explainable artificial intelligence can improve prediction accuracy over linear models and reveal great insights into outcomes like mortality. This paper comprehensively analyzes all-cause mortality by explaining complex machine learning models. Methods: We propose the IMPACT framework that uses XAI technique to explain a state-of-the-art tree ensemble mortality prediction model. We apply IMPACT to understand all-cause mortality for 1-, 3-, 5-, and 10-year follow-up times within the NHANES dataset, which contains 47,261 samples and 151 features. Results: We show that IMPACT models achieve higher accuracy than linear models and neural networks. Using IMPACT, we identify several overlooked risk factors and interaction effects. Furthermore, we identify relationships between laboratory features and mortality that may suggest adjusting established reference intervals. Finally, we develop highly accurate, efficient and interpretable mortality risk scores that can be used by medical professionals and individuals without medical expertise. We ensure generalizability by performing temporal validation of the mortality risk scores and external validation of important findings with the UK Biobank dataset. Conclusions: IMPACT's unique strength is the explainable prediction, which provides insights into the complex, non-linear relationships between mortality and features, while maintaining high accuracy. Our explainable risk scores could help individuals improve self-awareness of their health status and help clinicians identify patients with high risk. IMPACT takes a consequential step towards bringing contemporary developments in XAI to epidemiology.
◌ CITATION ONLY
Full text is not openly licensed for redistribution here. Read it at the source:
Provenance
- Source
- OpenAlex
- DOI
- 10.1038/s43856-022-00180-x
- Canonical
- link ↗
- Fetched
- 2026-10-16 MST
Cite this
APA
Qiu, W., Chen, H., Dincer, A.B., Lundberg, S., Kaeberlein, M.R., & Lee, S. (2022). Interpretable machine learning prediction of all-cause mortality. <em>Communications Medicine</em>. https://doi.org/10.1038/s43856-022-00180-x
Vancouver
Qiu W, Chen H, Dincer AB, Lundberg S, Kaeberlein MR, Lee S. Interpretable machine learning prediction of all-cause mortality. Communications Medicine. 2022. doi:10.1038/s43856-022-00180-x.
BibTeX
@article{wei2022Interp,
title = {Interpretable machine learning prediction of all-cause mortality},
author = {Wei Qiu and Hugh Chen and Ayse B. Dincer and Scott Lundberg and Matt R. Kaeberlein and Su‐In Lee},
journal = {Communications Medicine},
year = {2022},
doi = {10.1038/s43856-022-00180-x},
}
Research neighborhood
References, citing works, and semantically nearest findings. Click a node to open it.
Related findings
Kidney Diseases 2024
Open access · CC-BY
A Klotho-Based Machine Learning Model for Prediction of both Kidney and Cardiovascular Outcomes in Chronic Kidney Disease
Aging 2023
Open access · OA
Potential of Urolithin A to improve joint health
GeroScience 2024
Open access · CC-BY
Associations between epigenetic aging and diabetes mellitus in a Swedish longitudinal study
Frontiers in aging neuroscience 2026
Citation only
Atherosclerotic cardiovascular disease risk is associated with AT(N)V neuroimaging biomarkers and cognitive impairment in the health and aging brain study-health disparities cohort.
BMC Medicine 2021
Open access · CC-BY
Relationship between diet quality scores and the risk of frailty and mortality in adults across a wide age spectrum
Nutrition & Metabolism 2026
Open access · CC-BY