Detection of abnormalities within the inner ear is a challenging task that, if automated, could provide support for the diagnosis and clinical management of various otological disorders. Inner ear malformations are rare and present great anatomical variation, which challenges the design of deep learning frameworks to automate their detection. We propose a framework for inner ear abnormality detection, based on a deep reinforcement learning model for landmark detection trained in normative data only. We derive two abnormality measurements: the first is based on the variability of the predicted configuration of the landmarks in a subspace formed by the point distribution model of the normative landmarks using Procrustes shape alignment and Principal Component Analysis projection. The second measurement is based on the distribution of the predicted Q-values of the model for the last ten states before the landmarks are located. We demonstrate an outstanding performance for this implementation on both an artificial (0.96 AUC) and a real clinical CT dataset of various malformations of the inner ear (0.87 AUC). Our approach could potentially be used to solve other complex anomaly detection problems.
ochlear Implantation (CI) is an effective and doable alternative to restore hearing in cases of bilateral severe to profound hearing loss in patients who do not benefit from using an individual sound amplification device. 1,2 According to various authors by far the most common causes of recurrent CI is the migration of the implant and/or extrusion, technical failure and implant misplacement of the electrode array in the spiral canal of the cochlea. In general, the rate of complications is low. In the presence of abnormalities of the inner ear the risk of improper administration of the active electrode into the cochlea increases. 3 This complication according to some authors is 0.17-2.12%. 4
Objectives to discuss a five-year experience in surgical treatment of temporal bone paragangliomas from the point of view of U. Fisch and D. Mattox classification modified by М. Sanna in 2013.
Material and methods. In the period from February 2015 till December 2020, we performed 130 operations to remove temporal bone paraganglioma. The examined and operated patients included 34 men and 96 women aged from 2 to 82 years. The paraganglioma types A, B and C were distributed as follows: type A in 22 patients (A1 12 cases, A2 10 cases); type B in 73 patients (B1 25 cases, B2 16 cases, B3 32 cases); type C in 35 patients (С1 10 cases, С2 12 cases, С3 7 cases, С4 5 cases)
Results. The evaluated results included the quality of tumor removal, the auditory function and the function of the facial nerve in relation to the size of the neoplasm, registered during the early and late postoperative periods. Based on the study data, we developed an algorithm of tactics of surgical treatment of patients with this type of temporal bone pathology aimed to avoid damage to the vital structures of the lateral skull base.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.