GeHirNet: A Sex-Aware Hierarchical Model for Fair and Accurate Voice Pathology Classification

Jan 1, 2026·

Fan Wu

Kaiqi Zhao

Elgar Fleisch

Filipe Barata

· 0 min read

Cite

Abstract

GeHirNet is a sex-aware hierarchical deep learning model for voice pathology classification, achieving state-of-the-art performance (97% accuracy, 95% MCC) across 1,800+ patients.

Publication

IEEE International Conference on Healthcare Informatics

Last updated on Jan 1, 2026

Voice Machine Learning Healthcare Fairness

Authors

Fan Wu

AI Researcher

← Vocal Prognostic Digital Biomarkers in Monitoring Chronic Heart Failure: A Longitudinal Observational Study Apr 1, 2026

Leveraging Machine Learning for Developing Biological and Digital Biomarkers: Studies in Neurological Disorder Diagnostics, Circadian Rhythm Monitoring, and Heart Failure Prognostics Sep 30, 2025 →