Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.20341 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917355357995008 |
|---|---|
| author | Rachidi, Salma Bozorgpanah, Aso Fey, Eric Jung, Alexander |
| author_facet | Rachidi, Salma Bozorgpanah, Aso Fey, Eric Jung, Alexander |
| contents | Machine learning (ML) promises better clinical decision-making, yet opaque model behavior limits the adoption in healthcare. We propose two novel regularization techniques for ensuring the interpretability of ML models trained on real-world data. In particular, we consider the prediction of five-year survival for multiple myeloma patients using clinical data from Helsinki University Hospital. To ensure the interpretability of the trained models, we use two alternative constructions for a penalty term used for regularization. The first one penalizes deviations from the predictions obtained from an interpretable logistic regression method with two manually chosen features. The second construction requires consistency of model predictions with the revised international staging system (R-ISS). We verify the usefulness of the proposed regularization techniques in numerical experiments using data from 812 patients. They achieve an accuracy up to 0.721 on a test set and SHAP values show that the models rely on the selected important features. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_20341 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Interpretable Multiple Myeloma Prognosis with Observational Medical Outcomes Partnership Data Rachidi, Salma Bozorgpanah, Aso Fey, Eric Jung, Alexander Machine Learning Machine learning (ML) promises better clinical decision-making, yet opaque model behavior limits the adoption in healthcare. We propose two novel regularization techniques for ensuring the interpretability of ML models trained on real-world data. In particular, we consider the prediction of five-year survival for multiple myeloma patients using clinical data from Helsinki University Hospital. To ensure the interpretability of the trained models, we use two alternative constructions for a penalty term used for regularization. The first one penalizes deviations from the predictions obtained from an interpretable logistic regression method with two manually chosen features. The second construction requires consistency of model predictions with the revised international staging system (R-ISS). We verify the usefulness of the proposed regularization techniques in numerical experiments using data from 812 patients. They achieve an accuracy up to 0.721 on a test set and SHAP values show that the models rely on the selected important features. |
| title | Interpretable Multiple Myeloma Prognosis with Observational Medical Outcomes Partnership Data |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2603.20341 |