Automatic Prediction of Conductive Hearing Loss Using Video Pneumatic Otoscopy and Deep Learning Algorithm

Abstract

Objectives: Diseases of the middle ear can interfere with normal sound transmission, which results in conductive hearing loss. Since video pneumatic otoscopy (VPO) findings reveal not only the presence of middle ear effusions but also dynamic movements of the tympanic membrane and part of the ossicles, analyzing VPO images was expected to be useful in predicting the presence of middle ear transmission problems. Using a convolutional neural network (CNN), a deep neural network implementing computer vision, this preliminary study aimed to create a deep learning model that detects the presence of an air-bone gap, conductive component of hearing loss, by analyzing VPO findings. Design: The medical records of adult patients who underwent VPO tests and pure-tone audiometry (PTA) on the same day were reviewed for enrollment. Conductive hearing loss was defined as an average air-bone gap of more than 10 dB at 0.5, 1, 2, and 4 kHz on PTA. Two significant images from the original VPO videos, at the most medial position on positive pressure and the most laterally displaced position on negative pressure, were used for the analysis. Applying multi-column CNN architectures with individual backbones of pretrained CNN versions, the performance of each model was evaluated and compared for Inception-v3, VGG-16 or ResNet-50. The diagnostic accuracy predicting the presence of conductive component of hearing loss of the selected deep learning algorithm used was compared with experienced otologists. Results: The conductive hearing loss group consisted of 57 cases (mean air-bone gap = 25 ± 8 dB): 21 ears with effusion, 14 ears with malleus-incus fixation, 15 ears with stapes fixation including otosclerosis, one ear with a loose incus-stapes joint, 3 cases with adhesive otitis media, and 3 ears with middle ear masses including congenital cholesteatoma. The control group consisted of 76 cases with normal hearing thresholds without air-bone gaps. A total of 1130 original images including repeated measurements were obtained for the analysis. Of the various network architectures designed, the best was to feed each of the images into the individual backbones of Inception-v3 (three-column architecture) and concatenate the feature maps after the last convolutional layer from each column. In the selected model, the average performance of 10-fold cross-validation in predicting conductive hearing loss was 0.972 mean areas under the curve (mAUC), 91.6% sensitivity, 96.0% specificity, 94.4% positive predictive value, 93.9% negative predictive value, and 94.1% accuracy, which was superior to that of experienced otologists, whose performance had 0.773 mAUC and 79.0% accuracy on average. The algorithm detected over 85% of cases with stapes fixations or ossicular chain problems other than malleus-incus fixations. Visualization of the region of interest in the deep learning model revealed that the algorithm made decisions generally based on findings in the malleus and nearby tympanic membrane. Conclusions: In this preliminary study, the deep lear...

Advances in Image‐Based Artificial Intelligence in Otorhinolaryngology–Head and Neck Surgery: A Systematic Review

“…Some showed the superiority of AI over nonspecialist physicians 26 or otolaryngologists 71 in narrowly defined fields. Some studies used a single AI algorithm, while others combined 2 or more algorithms to solve clinical problems 25,38,63,74,75,79 . The data sources were used to train, validate, and test AI models, including single‐ and multiple‐center databases, 52,81 and public databases 25,85 …”

“…Since video pneumatic otoscopy (VPO) findings reveal not only the presence of middle ear effusions but also dynamic movements of the tympanic membrane and part of the ossicles, analyzing VPO images was expected to be useful in predicting the presence of middle ear transmission problems. Therefore, Byun et al 79 used Inception‐v3 to develop an AI model with 1130 VPO images to predict conductive hearing loss. Their models in predicting conductive hearing loss had 97.2% AUC and 94.1% accuracy, superior to that of experienced otologists, whose performance had an AUC of 77.3% and an average accuracy of 79.0%.…”

“…Recent studies have applied AI to otological imaging in various clinical contexts (Supplemental File 3, available online). These studies combined AI with otoscopy, [22][23][24][25][26]31,32,38,48,52,57,68,[74][75][76]79,81,84,85,93,95 computed tomography (CT), 30,36,41,42,50,62,63,70,71,88 and magnetic resonance imaging (MRI). 43,73,78 Most studies have focused on the image-based otoscopic diagnosis and automated segmentation of temporal bone CT for classifying normal and abnormal mastoid air cells.…”

“…Some studies used a single AI algorithm, while others combined 2 or more algorithms to solve clinical problems. 25,38,63,74,75,79 The data sources were used to train, validate, and test AI models, including single-and multiple-center databases, 52,81 and public databases. 25,85 Viscaino et al 26 used 22,000 otoscopic images to train a model that classifies 4 middle and external ear conditions: normal, chronic otitis media, otitis media with effusion, and earwax plugs.…”

Deep Learning Models for Predicting Hearing Thresholds Based on Swept-Tone Stimulus-Frequency Otoacoustic Emissions

“…With CNNs, the performance of the DL model in classifying chronic otitis media versus normal from temporal bone computed tomography scans was comparable or superior to that of clinical experts (Wang et al 2020b). A recent attempt to implement CNNs in the detection of conductive hearing loss via video pneumatic otoscopy images also achieved excellent performance (Byun et al 2022). However, DL techniques have not yet been applied to exploit the potential of OAEs in predicting hearing thresholds.…”