Objective comparison between perforation and hearing loss

Image-Based Artificial Intelligence Technology for Diagnosing Middle Ear Diseases: A Systematic Review

Otolaryngological diagnoses, such as otitis media, are traditionally performed using endoscopy, wherein diagnostic accuracy can be subjective and vary among clinicians. The integration of objective tools, like artificial intelligence (AI), could potentially improve the diagnostic process by minimizing the influence of subjective biases and variability. We systematically reviewed the AI techniques using medical imaging in otolaryngology. Relevant studies related to AI-assisted otitis media diagnosis were extracted from five databases: Google Scholar, PubMed, Medline, Embase, and IEEE Xplore, without date restrictions. Publications that did not relate to AI and otitis media diagnosis or did not utilize medical imaging were excluded. Of the 32identified studies, 26 used tympanic membrane images for classification, achieving an average diagnosis accuracy of 86% (range: 48.7-99.16%). Another three studies employed both segmentation and classification techniques, reporting an average diagnosis accuracy of 90.8% (range: 88.06-93.9%). These findings suggest that AI technologies hold promise for improving otitis media diagnosis, offering benefits for telemedicine and primary care settings due to their high diagnostic accuracy. However, to ensure patient safety and optimal outcomes, further improvements in diagnostic performance are necessary.

EAR-UNet: A deep learning-based approach for segmentation of tympanic membranes from otoscopic images

This paper presents a method for automatic segmentation of tympanic membranes (TMs) from video-otoscopic images based on deep fully convolutional neural network. Built upon the UNet architecture, the proposed EAR scheme is based on three main paradigms: EfficientNet for the encoder, Attention gate for the skip connection path, and Residual blocks for the decoder. The paper also introduces a new loss function term for the neural networks to perform segmentation tasks. Particularly, we propose to integrate EfficientNet-B4 into the encoder part of the UNet. In addition, the decoder part of the proposed network is constructed based on residual blocks from ResNet architecture. By this way, the proposed approach could take advantages of the EfficientNet and ResNet architectures such as preserving efficient reception field size for the model and avoiding overfitting problem. In addition, in the skip connection path, we employ the attention gate that can handle the varieties in shapes and sizes of interested objects, which are common issues in TM regions. Moreover, for network training, we proposed a new loss function term based on the shape distance between predicted and ground truth masks, and exploited the stochastic weight averaging to avoid being trapped in local minima. We evaluate the proposed approach on a TM dataset which includes 1012 otoscopic images from patients diagnosed with and without otitis media. Experimental results show that the proposed approach achieves high segmentation performance with the average Dice similarity coefficient of 0.929, without any pre- or post-processing steps, that outperforms other state-of-the-art methods.

Impact of Site, Size and Duration of Tympanic Membrane Perforation on Hearing Loss and Postsurgical Outcome

This paper aims to assess correlation of site, size and duration of tympanic membrane perforation with hearing loss using pure tone audiogram and surgical outcome in terms of above parameters. The present study was conducted on 100 patients in Department of ENT and HNS, SMGS Hospital, Government Medical College Jammu during a time period of November 2018 to October 2019. All the patients with age 15-60 years who presented with tympanic membrane (pars tensa) perforation were included in the study. According to the size of perforation, mean pure tone threshold in group I was 20.87 ± 3.86 dB, in group II was 26.45 ± 6.08 dB and in group III was 32.6 ± 5.56 dB. The difference in hearing threshold between all the three groups was significant statistically. In terms of site, group E had maximum hearing threshold (34.67 ± 4.20 dB), followed by group B (32.71 ± 5.88 dB). Group A had the lowest hearing threshold of 24.99 ± 6.21 dB. The difference between hearing thresholds of group B perforations and group A perforations was statistically significant (p < 0.05). However, the difference between group E and group B was insignificant. This study has shown significant correlation between the size and the site of the perforation to the degree of hearing loss. The bigger the perforation, the greater the hearing loss. The central perforations were associated with more hearing loss than posterior perforations, thus refuting the hypothesis that site and size of a tympanic membrane perforation does not affect the degree of conductive hearing loss. This study did not show any correlation between duration of disease and degree of hearing loss. Surgical and audiometric results obtained in this study can be accepted as satisfactory and as expected by the literature.

A Study of Correlation Between Tympanic Membrane Perforation Size With Hearing Loss in Patients With Inactive Mucosal Chronic Otitis Media

PURPOSE

This study is directed towards establishing a correlation between the objective measurement of tympanic membrane (TM) perforation size with hearing loss in patients with inactive mucosal chronic otitis media (COM), and to compare the otoscopic findings of the TM perforation vis-a-vis the results obtained by image analysis of the endoscopic image of the perforation.

MATERIALS AND METHODS

An observational, prospective and hospital based study was performed on 107 ears (78 patients) with inactive mucosal chronic otitis media. Objective measurement of the TM perforation size was done by capturing the images of the perforation on ImageJ analysis software. A correlation was established between the hearing loss and the perforation with respect to its size and location. The accuracy of the otoscopic finding of the perforation was established with respect to the endoscopic image of the perforation.

RESULTS

The extent of the hearing loss was seen to increase with increasing size of the TM perforation (r = 0.435, p < 0.0001). Posterior perforations had a 12% greater hearing loss than anterior perforations but this was significant only at the frequency of 500 Hz (p = 0.02). There was a positive correlation between the otoscopic estimation of the size of the perforation with the endoscopically obtained image.

CONCLUSION

Hearing loss in chronic otitis media is directly proportional to the size of the TM perforation and posteriorly based perforations have worse audiometric thresholds at lower frequencies (500 Hz).

Determinants of hearing loss severity in tympanic membrane perforations in a sub-Saharan African setting

BACKGROUND

Factors specific to sub-Saharan Africa could have an impact on tympanic membrane perforation characteristics. Obtaining precise information on these characteristics and determinants of hearing loss severity would enable better management.

OBJECTIVE

To determine the relationship between characteristics of tympanic membrane perforation and subsequent hearing impairment.

METHODS

A cross-sectional study of consenting patients presenting with tympanic membrane perforation was conducted. They were examined using otoendoscopy with a digital camera to obtain precise measurements, followed by pure tone audiometry.

RESULTS

Eighty-six cases of tympanic membrane perforation were included. Mean tympanic membrane perforation proportion was 34.1 ± 18.4 per cent. Medium-sized tympanic membrane perforations were predominant (47.7 per cent). Median tympanic membrane perforation duration was 20 years. Tympanic membrane perforation size was found to be a predictor of hearing loss severity (odds ratio = 2.5, 95 per cent confidence interval = 1.02-6.13, p = 0.04).

CONCLUSION

Tympanic membrane perforation size was a predictor of hearing loss severity in our setting. Site, duration and aetiology seem to have no impact on hearing loss severity.