Age-Conditioned Evaluation of Handcrafted Polysomnography Models for Cognitive-Impairment Prediction

Satwik Kapavarapu N. Hoang Andrei Malagusan

· 0 citations · 10 references

TL;DR

A CPU-oriented system that represented each study with 830 handcrafted demographic, physiological, and annotation-derived features indicated that all-pair discrimination did not persist among age-comparable participants and identifies age-conditioned cross-source generalization as the principal limitation of the submitted approach.

Semantic Scholar

View source

Similar papers

Cross-Site Prediction of Future Cognitive Impairment from Polysomnography Using Hybrid Physiological Representations

Alireza Khorramfard, Naimahmed Nesaragi, Daniel Teichmann et al. · 0 citations

Site-Aware Stacking for Cross-Center Cognitive Risk Modeling From Overnight Polysomnography

Yuxuan Xu · 0 citations

Frozen SleepFM and engineered polysomnographic features for prediction of subsequent cognitive impairment diagnoses

Karen K. Wurzel, D. Kaftan, M. Du Pont et al. · 0 citations

Age-Residualized Physiological Biomarkers and Gradient-Boosted Trees for Predicting Cognitive Impairment from Polysomnography

This entry extracts a bank of 136 literature-motivated physiological features from the measurements, stratified by sleep stage, and compute an age-residual feature for a curated set of 23 biomarkers with established age trends.

Paul D. Fischer, Simon Boehi, S. El Hadramy et al. · 0 citations

Autonomous Agentic Experimentation for Predicting Cognitive Impairment from Polysomnography

The George B. Moody PhysioNet Challenge 2026 asks for a future cognitive-impairment diagnosis from one overnight polysomnogram, ranked by an age-conditioned AUROC over age-matched pairs, and reports a process and the model it selected, separately.

Gyula Bohus, József Nagy-Bozsoky, Klaudia Vivien et al. · 0 citations

SIREN: A Site-Invariant Rank Ensemble for Predicting Future Cognitive Impairment from Polysomnography

María Muñoz Pérez, Alejandro Hinojosa, Lorena Gallego-Vi ̃nar'As et al. · 0 citations