A Machine-Learned Comorbidity Index
Quick Answer
This paper shows that The Machine-Learned Comorbidity Index (MLCI) improves upon traditional comorbidity scores by capturing nonlinear risk-outcome relationships, outperforming established benchmarks in electronic health record datasets.
Quick Take
MLCI maximizes the normalized Hilbert-Schmidt Independence Criterion (nHSIC) for better patient stratification across multiple clinical outcomes.
Key Points
- MLCI addresses limitations of traditional scores like Charlson and Elixhauser.
- It captures nonlinear relationships between comorbidities and clinical outcomes.
- Empirical tests show MLCI outperforms strong baseline models in EHR datasets.
- The model is grounded in a theoretical framework for unified patient ordering.
- MLCI enhances risk adjustment and patient stratification in healthcare.
Paper Resources
Article Excerpt
From source RSS / original summaryarXiv:2606. 17450v1 Announce Type: new Abstract: Traditional comorbidity scores (e. g. , Charlson and Elixhauser) are widely used for risk adjustment and patient stratification, but they have two key limitations: (i) they are largely mortality-centric and do not align well with other clinical outcomes, and (ii) their linear, rule-based structure cannot capture nonlinear, outcome-specific risk relationships.
We propose a Machine-Learned Comorbidity Index (MLCI) that maps diagnosis codes to a single scalar by maximizing the normalized Hilbert-Schmidt Independence Criterion (nHSIC) between the learned score and multiple clinical outcomes. MLCI captures nonlinear risk-outcome dependence and is supported by a theory that characterizes when a unified, informative admission-level ordering can be achieved across outcomes.
Empirical results on multiple benchmark electronic health record (EHR) datasets show that MLCI outperforms strong baselines across multiple evaluation metrics.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.