A rapid, machine-learning readout of cochlear neural degeneration that transfers from animals to humans
- 1University of Pittsburgh
Abstract
Standard hearing tests miss cochlear neural degeneration (CND), the loss of synapses between inner hair cells and auditory nerve fibers that may underlie unexplained hearing difficulty in roughly one in ten adults. Envelope following responses (EFRs) can index this neural loss noninvasively, but conventional protocols are too slow for clinical use. We tested a dynamic amplitude-modulated (dAM) EFR that sweeps the full modulation spectrum in a single brief stimulus. In Mongolian gerbils with histologically verified CND, dAM EFRs revealed selective deficits at fast modulation rates without threshold elevation. A machine-learning classifier distinguished young from middle-aged animals with high accuracy; additionally, its most informative feature, power near 400-500 Hz, tracked synapse counts, indicating a readout of cochlear neural health rather than age. Applied without retraining to 56 human listeners, the gerbil-trained classifier separated age groups above chance. dAM EFRs thus offer a rapid, objective and clinically viable marker of CND.
Related
- Integrated transcriptomic, machine learning, and network pharmacology analyses reveal shared inflammatory biomarkers between intervertebral disc degeneration and osteoporosis
- Cross-family learning transfers thermal adaptation between enzymes
- Memories that remain: Image memorability transfers to scene descriptions
- What Transfers Across Brains Is Not What Stays Within Individuals: Evidence from Bilingual Language States
- NSD2 deficiency disrupts the epigenetic landscape of cochlear hair cells to enhance ECM expression and to perturb cochlear hair cell function
BibTeX
@misc{andrino2026rapid,
title = {{A rapid, machine-learning readout of cochlear neural degeneration that transfers from animals to humans}},
author = {L. S. Andrino and R. Gall and J. Klara and M. Akcakaya and E. L. Bartlett and A. Parthasarathy},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.24.754199},
url = {https://www.biorxiv.org/content/10.64898/2026.09.24.754199v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - A rapid, machine-learning readout of cochlear neural degeneration that transfers from animals to humans
AU - L. S. Andrino
AU - R. Gall
AU - J. Klara
AU - M. Akcakaya
AU - E. L. Bartlett
AU - A. Parthasarathy
PY - 2026
DA - 2026/09/24
PB - bioRxiv
DO - 10.64898/2026.09.24.754199
UR - https://www.biorxiv.org/content/10.64898/2026.09.24.754199v1
AB - Standard hearing tests miss cochlear neural degeneration (CND), the loss of synapses between inner hair cells and auditory nerve fibers that may underlie unexplained hearing difficulty in roughly one in ten adults. Envelope following responses (EFRs) can index this neural loss noninvasively, but conventional protocols are too slow for clinical use. We tested a dynamic amplitude-modulated (dAM) EFR that sweeps the full modulation spectrum in a single brief stimulus. In Mongolian gerbils with histologically verified CND, dAM EFRs revealed selective deficits at fast modulation rates without threshold elevation. A machine-learning classifier distinguished young from middle-aged animals with high accuracy; additionally, its most informative feature, power near 400-500 Hz, tracked synapse counts, indicating a readout of cochlear neural health rather than age. Applied without retraining to 56 human listeners, the gerbil-trained classifier separated age groups above chance. dAM EFRs thus offer a rapid, objective and clinically viable marker of CND.
ER -