Computer-assisted ABR interpretation using the automatic construction of the latency-intensity curve

Area of Science:

• Audiology
• Biomedical Engineering
• Signal Processing

Background:

• Accurate measurement of auditory brainstem response (ABR) is crucial for hearing diagnostics.
• Traditional methods for constructing latency intensity (LI) curves can be time-consuming and prone to errors.
• Automating LI curve construction can improve efficiency and reliability in audiological assessments.

Purpose of the Study:

• To develop a novel computerized technique for the automatic construction of the latency intensity (LI) curve.
• To improve the accuracy of response detection and latency measurements in auditory brainstem response (ABR) analysis.
• To validate the performance of the automatic LI curve construction method against expert evaluation.

Main Methods:

• A pattern recognition approach utilizing a priori information from audiograms and physiological data.
• Analysis of recordings at various stimulus intensities and extraction of single auditory brainstem response (ABR) waveforms.
• Implementation of an algorithm for automatic wave recognition and latency measurement.

Main Results:

• The automated technique successfully prevented misrecognition errors in response detection and latency measurements.
• Automatic recognition of auditory brainstem response (ABR) waves coincided with ear, nose, and throat (ENT) specialist recognition in over 90% of cases.
• For wave V, the average deviation between the automated algorithm and ENT specialist for response thresholds was 5 dB, and for latencies was 0.05 ms.

Conclusions:

• The developed computerized technique offers a reliable and accurate method for automatic LI curve construction.
• This automated approach significantly reduces errors in ABR analysis, enhancing diagnostic precision.
• The high degree of agreement with specialist recognition validates the clinical utility of the proposed method.