# Temporal Sepsis Modeling: a Relational and Explainable-by-Design Framework

arXiv: https://arxiv.org/abs/2601.21747 · HTML: https://mlsanity.com/paper/2601.21747/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 35 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Builds on (most cited first, 7 of 7)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Explanation in Artificial Intelligence: Insights from the Social Sciences](https://mlsanity.com/paper/1706.07269/) (2017)
- [Even if Explanations: Prior Work, Desiderata & Benchmarks for Semi-Factual XAI](https://mlsanity.com/paper/2301.11970/) (2023)
- [An Improved Multi-Output Gaussian Process RNN with Real-Time Validation for Early Sepsis Detection](https://mlsanity.com/paper/1708.05894/) (2017)
- [Fractional Naive Bayes (FNB): non-convex optimization for a parsimonious weighted selective naive Bayes classifier](https://mlsanity.com/paper/2409.11100/) (2024)
- [An Efficient Shapley Value Computation for the Naive Bayes Classifier](https://mlsanity.com/paper/2307.16718/) (2023)
- [Viewing the process of generating counterfactuals as a source of knowledge: a new approach for explaining classifiers](https://mlsanity.com/paper/2309.04284/) (2023)

## Abstract (arXiv)

Sepsis remains one of the most complex and heterogeneous syndromes in intensive care. While deep learning models achieve competitive performance in early sepsis prediction, their decision processes often remain difficult to interpret clinically, and explainability is typically added only through post-hoc methods. We propose an explainable-by-design framework based on a relational approach: temporal EHR data are represented in a relational schema, flattened via MDL-based propositionalisation into compact human-readable features, and classified using a selective Fractional Naive Bayes classifier. Evaluated on MIMIC-III (3,940 patients, 10-fold cross-validation), our approach achieves AUC = 0.983 - competitive with XGBoost (0.985) and CatBoost (0.985), and superior to LSTM (0.945) - with only 98 selected variables and a 1 MB model footprint. Unlike post-hoc methods, interpretability is native and fourfold: univariate, global, local, and counterfactual.