Abstract
Predictive modeling for clinical tabular data is central to clinical decision support and requires not only strong predictive performance but also transparent decision logic. While deep learning and tree-based ensemble methods can achieve high accuracy, their black-box nature remains a significant obstacle to clinical deployment. This challenge is exacerbated by common characteristics of medical data, including limited sample sizes, severe class imbalance, and feature evolution due to changes in diagnostic criteria and clinical documentation.
To tackle these issues, we propose Medical Heuristic Learning (MHL), an instantiation of the learning-beyond-gradients paradigm for clinical tabular prediction. Rather than relying on neural network weight updates, MHL employs a large language model (LLM)-driven workflow that integrates statistical probes, medical knowledge probes, rule synthesis, and code-level iterative refinement to optimize a deterministic and executable decision system. The resulting model is expressed not as opaque parameters, but as versioned pure-Python decision rules that are explicitly interpretable, fully auditable, and clinically grounded.
MHL also supports continual learning by starting from previously validated rules and iteratively revising them using updated feature information under data drift or feature evolution. Comprehensive experiments on medical datasets show that MHL achieves performance comparable to state-of-the-art methods while maintaining strong behavior in small-sample and highly imbalanced settings. The results further indicate that this explicit rule update mechanism can help alleviate catastrophic forgetting under feature evolution. Overall, these findings suggest that non-gradient-based heuristic systems offer a transparent and adaptable alternative for high-stakes clinical decision support.
Blogger's Review: The introduction of MHL effectively addresses the issues of transparency and auditability in medical data, making clinical decisions both efficient and trustworthy. This framework not only supports high-performance decision-making but also lays a strong theoretical and practical foundation for future applications of AI in healthcare, showcasing the immense potential of modern technology in this field.