Related Experiment Video
Updated: Jun 15, 2025
The Power of Interstimulus Interval for the Assessment of Temporal Processing in Rodents
Published on: April 19, 2019
An interpretable tinnitus prediction framework using gap-prepulse inhibition in auditory late response and
Iqram Hussain1, Chiheon Kwon2, Tae-Soo Noh3
1Institute of Medical and Biological Engineering, Medical Research Center, Seoul National University College of Medicine, Seoul 03080, Republic of Korea; Department of Anesthesiology, Weill Cornell Medicine, Cornell University, New York, NY 10065, USA.
This study introduces a new tinnitus diagnostic method using auditory late response (ALR) and electroencephalogram (EEG) data. The interpretable model accurately identifies tinnitus by analyzing gap-prepulse inhibition (GPI) and EEG spectral features.
Area of Science:
- Neuroscience
- Auditory Neuroscience
- Medical Diagnostics
Background:
- Tinnitus is a neuropathological condition characterized by phantom ear sounds.
- Current diagnostic methods for tinnitus are often subjective and complex.
- There is a need for objective and interpretable diagnostic tools for tinnitus.
Purpose of the Study:
- To propose an interpretable tinnitus diagnostic framework.
- To utilize auditory late response (ALR) and electroencephalogram (EEG) data.
- To incorporate the gap-prepulse inhibition (GPI) paradigm into tinnitus diagnosis.
Main Methods:
- Collected EEG and ALR data from tinnitus patients and controls.
- Analyzed EEG spectral and ALR features in response to stimuli with embedded gaps.
- Developed an interpretable machine learning model using ALR and EEG metrics.
Main Results:
- Achieved 90% accuracy in tinnitus identification with an AUC of 0.89.
- Identified gap-embedded ALR (GPI ratio of N1-P2) and EEG spectral ratio as key diagnostic metrics.
- Provided personalized prediction explanations for tinnitus diagnosis.
Conclusions:
- Deficits in GPI and specific EEG alpha-beta activity are promising indicators for tinnitus risk.
- The proposed framework offers a potential screening tool for tinnitus.
- Findings align with existing clinical insights in hearing research.
