Automatic Analysis of the Patient´s Conscious Responses to the Emission of Auditory Stimuli during the Performance of an Audiometry

This paper presents a methodology for the automatic detection of human conscious responses to different auditory stimuli during the performance of an audiogram. An audiogram is the standard way of analyze the hearing of a patient in order to diagnose hearing loss. From a video sequence captured during the hearing evaluation performed by an expert to a patient, the methodology will be able to detect the instants in which the expert is sending auditory stimuli and when the patient responds consciously to them by raising his hand in order to indicate that he has heard them. The proposed method has been tested on a dataset of almost 7000 frames from three different video sequences obtaining very encouraging results in the hand raise detection as a response to the auditory stimuli. Finally, an analysis of the patient´s speed reaction to stimuli is discussed.