Rule-based system for interpretation of brainstem auditory evoked potentials

A preliminary version of an expert decision-support system for interpretation of brainstem auditory evoked potentials is described. The design combines an analytical signal-processing module with a rule module that incorporates the kinds of heuristic criteria used by human interpreters. The system determines whether a sensory response is present in the waveform; if so, it identifies peak V, the most prominent peak. The system performed well when a response was present, but it was not as effective in interpreting waveforms with no response present. Improvement in analytical signal detection and expansion of the heuristic database could provide performance acceptable for clinical use.<>