Despite the variety of etiological factors, cochlear implantation (CI) remains the only effective method for the rehabilitation of the patients presenting with total deafness. The aim of this study was the enhancement of the efficiency of selection of the candidates for CI, the improvement of the quality of rehabilitation of the patients with cochlear implants, and the determination of the prognostic criteria for clinical trials. Patients and methods. (CI). The decision-making support system (DMSS) based on the artificial neural networks (ANNs) has been created to enhance the efficiency of rehabilitation of the patients with cochlear implants and increase the effectiveness of the selection of candidates for cochlear implantation. The results of the children’s rehabilitation after CI have been analyzed by using a mathematical model of artificial neural networks (Kohonen layer). The basis for the assessment of ANNs was formed by the results of the observations of audioverbal perception in 110 patients aged from 6 months to 17 years. The initial data were the average values obtained with the use of the Russian-language version of the Nottingham children’s implant profile’s test T1 — T3. The testing was performed before CI and 3, 6, 12, 18, and 24 months after it. Main results. The work yielded the four-cluster data structure. It made it possible to estimate the effectiveness of the clinical trials in selected classes depending on the etiology of the disease, the age of the patients, and their experience with the application of hearing aids. The reliable estimation of the dynamics of auditory perception at the stage of rehabilitation and prognosis of the outcomes of CI made it possible to take additional preventive and therapeutic measures in the combination with complementary psychological and educational procedures.