Deep Hybrid Medical Vision System for Early Otitis Media Diagnosis using Otoscopic Imaging

Read full text Summarise with VIRAT
Authors: Zareena Begum, Martha Navya Sri, Md Afrin, Kotte Manesh, Manugonda Muneeshwari
Cited by: 0
References: 35

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Otitis media includes several common inflammatory conditions of the middle ear that can have severe complications if left untreated.Correctly identifying otitis media can be difficult and a screening system supported by machine learning would be valuable for this prevalent disease.The study presents a deep learning-based automated framework for diagnosing otitis media from otoscopic images, addressing the increasing global burden of ear infections, particularly among children, where the disease remains a leading cause of hearing impairment and frequent ENT consultations.Although digital otoscopic devices generate large volumes of clinical images, diagnosis still predominantly relies on manual visual inspection by otolaryngologists, making it subjective, time-consuming, and prone to inter-observer variability, delayed detection, and fatigue-related errors, especially in resource-limited settings.Conventional assessment involves identifying conditions such as Acute Otitis Media, Chronic Otitis Media, Myringosclerosis, Cerumen Impaction, and Normal cases through expert interpretation, which lacks standardization and scalability.To overcome these challenges, the proposed system integrates image preprocessing, deep feature extraction using DenseNet121, and classification using machine learning models including Nearest Centroid Classifier (NCC), K-Nearest Neighbors (KNN), and eXtreme Gradient Boosting (XGBoost), and proposed calibrated perceptron with Dense neural network (DNN) also called as ensemble Voting model to enhance predictive performance.The dataset is divided into training and testing subsets, and the models are evaluated using accuracy, precision, recall, F1-score, confusion matrix, ROC curves, and AUC metrics.Experimental results demonstrate that the Voting-based ensemble model achieves superior performance compared to individual classifiers, providing a reliable, efficient, and scalable solution for automated otitis media detection and clinical decision support.

Fields of Study & Subjects

Ear Surgery and Otitis Media Artificial Intelligence in Healthcare Smart Agriculture and AI

Publication details

Journal / source
Journal of Science Engineering Technology and Management Sciences
Volume
3
Issue
3
Pages
233–244
Publication date
1 March 2026
Language
EN