HQML-ICU: hybrid quantum-classical machine learning for ICU mortality prediction using MIMIC-IV electronic health records
Source check: Source checked. Not independent scientific confirmation.
Finding: In a retrospective single-center analysis of 38,624 MIMIC-IV ICU admissions, a hybrid quantum-classical model predicted in-hospital mortality with test AUROC 0.871 versus 0.858 for XGBoost (DeLong p=0.012), with the difference also reported in five-fold cross-validation. The quantum circuit ran in noiseless statevector simulation on classical hardware; external validation is needed.
Technology: A learned hybrid quantum-classical model (classical autoencoder, variational quantum circuit and classical output layer) predicts in-hospital mortality from first-day ICU data.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- A learned hybrid quantum-classical model (classical autoencoder, variational quantum circuit and classical output layer) predicts in-hospital mortality from first-day ICU data.
- Test and result
- Test-set comparison and five-fold cross-validation on a single-center MIMIC-IV cohort; authors describe noiseless classical statevector simulation.
- Limitation / still unknown
- Retrospective single-center cohort; simulated quantum circuit on classical hardware and no external validation.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1016/j.compbiomed.2026.109876
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary full text, including Methods and Results; publication date and first result-bearing version checked..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.