AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

A Machine-Learned Comorbidity Index

Suleman Baloch, Kishlay Jha, Alberto Maria Segre, Philip M. Polgreen, Bijaya Adhikari

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Statistical analysis of ML practice · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train score with multi-outcome nHSIC instead of fixed-weight clinical indexMLCI vs Elixhauser · MIMIC-IV MORTdistance correlation x10^2: 19.98 -> 54.80Improved
train score with multi-outcome nHSIC instead of fixed-weight clinical indexMLCI vs Elixhauser · MIMIC-III 30Mdistance correlation x10^2: 24.70 -> 37.39Improved
nHSIC objective instead of BCE for the same DeepSets architectureDeepSets encoder · MIMIC-IV MORTdistance correlation x10^2: 28.51 -> 54.80Improved
nHSIC objective instead of BCE for the same DeepSets architectureDeepSets encoder · MIMIC-IV ICUdistance correlation x10^2: 51.92 -> 61.97Improved
nHSIC objective instead of BCE for the same DeepSets architectureDeepSets encoder · MIMIC-IV LOSdistance correlation x10^2: 48.60 -> 51.15No clear effect
nHSIC objective instead of BCE for the same DeepSets architectureDeepSets encoder · MIMIC-III ICU (late transfer)distance correlation x10^2: 21.52 -> 18.55Hurt
nHSIC objective instead of BCEDeepSets encoder · MIMIC-III MORTmutual information x10^3: 73.69 -> 84.52Improved
nHSIC objective instead of BCEDeepSets encoder · MIMIC-IV ICUmutual information x10^2: 19.24 -> 19.01Mixed
use learned/ML score instead of CCICharlson · MIMIC-IV MORTdistance correlation x10^2: 12.59 -> 54.80Improved
add Elixhauser to Charlson as a summed scalarCCI · MIMIC-IV MORTdistance correlation x10^2: 12.59 -> 17.46Improved
rank-one projection of stacked label structure as shared directionMLCI theory diagnostic · MIMIC-IVrank-one energy share: 0.49No clear effect
compare best full-objective threshold with rank-one thresholdMLCI · MIMIC-IV vs MIMIC-IIIthreshold split agreement: jJ 4863 = jv 4863 (IV); jJ 6707 vs jv 11128 (III)Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy