NEURAL PREDICTORS OF LANGUAGE-SKILL OUTCOMES IN COCHLEAR IMPLANTATION PATIENTS

摘要

Machine-learning techniques are used to train a classifier to predict auditory and language skills improvement in a patient who is a candidate for cochlear implantation (CI). One or more images of portions of the patient's brain are obtained, and quantitative data is extracted that represents the composition of one or more brain areas related to auditory and/or cognitive processing. For training of the classifier, data is obtained for previous CI patients whose improvement in language skills has been measured. Once trained, the classifier can be used to predict a likely degree of improvement in a prospective CI patient's auditory and language skills.