Interpretable Multiple Myeloma Prognosis with Observational Medical Outcomes Partnership Data

TL;DR

This paper introduces two novel regularization techniques to make machine learning models for multiple myeloma prognosis more interpretable, using real-world clinical data and ensuring transparency in predictions.

Contribution

It proposes two new regularization methods that improve interpretability of ML models in healthcare, aligning predictions with clinical standards and simple models.

Findings

Achieved up to 72.1% accuracy on test data.

Models rely on clinically important features as shown by SHAP values.

Regularization techniques enhance interpretability without sacrificing accuracy.

Abstract

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.