Validated Methods for Synthesising Hearing Health Data for Machine Learning: A Comparative Study of KDE and VAE Approaches
Open Access
- 18 March 2026
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 16 (6) , 2917
Abstract
Hearing loss affects approximately 1.5 billion people globally, yet access to comprehensive audiometric datasets for research remains limited due to privacy constraints. Synthetic data generation offers a promising solution, enabling broader data sharing while preserving privacy. This study developed and validated two complementary approaches for synthesising audiometric data: Kernel Density Estimation (KDE) and Variational Autoencoders (VAE). Using the National Health and Nutrition Examination Survey (NHANES) dataset comprising 36,676 participants with comprehensive hearing assessments, we trained both generative models and evaluated synthetic data quality through a rigorous Train-on-Synthetic-Test-on-Real (TSTR) machine learning validation framework and blinded expert clinical assessment by two independent audiologists. The VAE approach achieved 86.3% utility for hearing loss prediction, as compared to the benchmark real data (Train-on-Real-Test-on-Real). Both methods demonstrated strong privacy preservation, with zero exact record matches and robust membership inference attack resistance. Statistical validation confirmed equivalence within clinically negligible margins (<1 dB HL) across all audiometric frequencies. Blinded assessment of 85 patient profiles by two independent expert audiologists revealed that VAE synthetic data achieved high clinical plausibility ratings, with 96.7% of VAE profiles rated as plausible, compared to 13.3% for KDE. Inter-rater reliability was moderate (Cohen’s weighted , ICC ), with 84.7% of ratings within one point, and both raters independently ranking VAE above real data above KDE. These findings establish validated methodologies for generating privacy-preserving synthetic audiometric data suitable for machine learning applications and clinical education, addressing a critical gap in hearing health research infrastructure.
Keywords
Funding Information
- National Institute for Health and Care Research (NIHR), University College London Hospitals Biomedical Research Centre (NIHR 203328 BRC965A/HH/AS/110390)
This publication has 47 references indexed in Scilit:
- Video saliency prediction via single feature enhancement and temporal recurrenceEngineering Applications of Artificial Intelligence, 2025
- Machine learning-based prediction of hearing loss: Findings of the US NHANES from 2003 to 2018Hearing Research, 2025
- Machine learning-based longitudinal prediction for GJB2-related sensorineural hearing lossComputers in Biology and Medicine, 2024
- Generation and evaluation of privacy preserving synthetic health dataNeurocomputing, 2020
- Primary Neural Degeneration in the Human Cochlea: Evidence for Hidden Hearing Loss in the Aging EarNeuroscience, 2019
- Declining Prevalence of Hearing Loss in US Adults Aged 20 to 69 YearsJAMA Otolaryngology–Head & Neck Surgery, 2017
- Drug‐induced ototoxicity: Mechanisms, Pharmacogenetics, and protective strategiesClinical Pharmacology & Therapeutics, 2017
- Prevalence of Hearing Loss and Differences by Demographic Characteristics Among US AdultsData From the National Health and Nutrition Examination Survey, 1999-2004Archives of internal medicine (1960), 2008
- Multivariate Density Estimation1992
- Density Estimation for Statistics and Data Analysis1986