![[2602.17364] A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.17364] A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data
Summary
This article presents CACTUS, a machine learning framework designed to enhance decision-making in clinical settings by ensuring feature stability and interpretability under incomplete data conditions.
Why It Matters
The framework addresses critical challenges in applying machine learning to biomedical data, such as robustness and interpretability, which are essential for trust in high-stakes medical decisions. By focusing on feature stability, CACTUS aims to improve the reliability of predictive models in clinical environments where data may be incomplete.
Key Takeaways
- CACTUS framework enhances feature stability and interpretability in machine learning models for clinical data.
- It demonstrates superior predictive performance while maintaining stability of key features under missing data conditions.
- The framework is designed to support trustworthy decision-making in high-stakes biomedical applications.
Computer Science > Machine Learning arXiv:2602.17364 (cs) [Submitted on 19 Feb 2026] Title:A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data Authors:Justyna Andrys-Olek, Paulina Tworek, Luca Gherardini, Mark W. Ruddock, Mary Jo Kurt, Peter Fitzgerald, Jose Sousa View a PDF of the paper titled A feature-stable and explainable machine learning framework for trustworthy decision-making under incomplete clinical data, by Justyna Andrys-Olek and Paulina Tworek and Luca Gherardini and Mark W. Ruddock and Mary Jo Kurt and Peter Fitzgerald and Jose Sousa View PDF HTML (experimental) Abstract:Machine learning models are increasingly applied to biomedical data, yet their adoption in high stakes domains remains limited by poor robustness, limited interpretability, and instability of learned features under realistic data perturbations, such as missingness. In particular, models that achieve high predictive performance may still fail to inspire trust if their key features fluctuate when data completeness changes, undermining reproducibility and downstream decision-making. Here, we present CACTUS (Comprehensive Abstraction and Classification Tool for Uncovering Structures), an explainable machine learning framework explicitly designed to address these challenges in small, heterogeneous, and incomplete clinical datasets. CACTUS integrates feature abstraction, interpretable classification, and systematic feature stab...