Deep Learning in Automated Region Proposal and Diagnosis of Chronic Otitis Media Based on Computed Tomography

Abstract

Objectives: The purpose of this study was to develop a deep-learning framework for the diagnosis of chronic otitis media (COM) based on temporal bone computed tomography (CT) scans. Design: A total of 562 COM patients with 672 temporal bone CT scans of both ears were included. The final dataset consisted of 1147 ears, and each of them was assigned with a ground truth label from one of the 3 conditions: normal, chronic suppurative otitis media, and cholesteatoma. A random selection of 85% dataset (n = 975) was used for training and validation. The framework contained two deep-learning networks with distinct functions: a region proposal network for extracting regions of interest from 2-dimensional CT slices; and a classification network for diagnosis of COM based on the extracted regions. The performance of this framework was evaluated on the remaining 15% dataset (n = 172) and compared with that of 6 clinical experts who read the same CT images only. The panel included 2 otologists, 3 otolaryngologists, and 1 radiologist. Results: The area under the receiver operating characteristic curve of the artificial intelligence model in classifying COM versus normal was 0.92, with sensitivity (83.3%) and specificity (91.4%) exceeding the averages of clinical experts (81.1% and 88.8%, respectively). In a 3-class classification task, this network had higher overall accuracy (76.7% versus 73.8%), higher recall rates in identifying chronic suppurative otitis media (75% versus 70%) and cholesteatoma (76% versus 53%) cases, and superior consistency in duplicated cases (100% versus 81%) compared with clinical experts. Conclusions: This article presented a deep-learning framework that automatically extracted the region of interest from two-dimensional temporal bone CT slices and made diagnosis of COM. The performance of this model was comparable and, in some cases, superior to that of clinical experts. These results implied a promising prospect for clinical application of artificial intelligence in the diagnosis of COM based on CT images.

Investigating the use of a two-stage attention-aware convolutional neural network for the automated diagnosis of otitis media from tympanic membrane images: a prediction model development and validation study

“…In recent years, Artificial Intelligence has been applied to medical image analysis to help clinical interpretation and medical diagnosis by image classification, segmentation and matching. [7][8][9][10] In particular, convolutional neural networks (CNNs) have been widely used and demonstrate good performance in the automated classification of medical images, including diabetic retinopathy detection, 9 10 skin cancer classification 11 12 and congenital cataract detection. 13 However, few machine-learning studies have been conducted in the field of otology for the automated diagnosis of ear diseases from otoscopic images.…”

The use of explainable artificial intelligence to explore types of fenestral otosclerosis misdiagnosed when using temporal bone high-resolution computed tomography

“…Previous diagnostic works based on deep learning paid less attention to otological diseases and more to the skin, breast, lung, liver, retina, esophagus, and colon (18,19,21,29,(31)(32)(33)(34)(35)(36)(37)(38). Some diagnostic works for ear disease achieved promising results such as the diagnosis of COM based on temporal bone CT scans using deep learning (22), diagnosis of secretory otitis media with machine learning algorithm (22), diagnosis of ear diseases based on otoscope images using ensemble deep models (39), and prediction of hearing and speech perception in children with cochlear implants using AI technology (40). However, there are no studies devoted to diagnosing fenestral OS lesions, which is the most likely ear disease to be misdiagnosed and has the most obvious therapeutic effect.…”

“…However, there are few studies on AI in the diagnosis and treatment of otological disease using CT images. To date, only one retrospective study based on temporal bone CT has described the application of AI in the distinction between COM and middle ear cholesteatoma (22). This is because the imaging manifestations of ear diseases are much smaller in size than those of other diseases.…”

Künstliche Intelligenz in der Hals-Nasen-Ohren-Heilkunde

“…Knocheneiterung und 79 %: Normalbefund). Ähnliche Ergebnisse erbrachte die Befundung des Bildmaterials durch Radiolog*innen und HNO-Ärzt*innen [ 31 ].…”