Enhanced Classification of Ear Disease Images Using Metaheuristic Feature Selection DOI Creative Commons
Furkancan Demircan, Murat Ekіncі, Zafer Cömert

et al.

Sakarya University Journal of Computer and Information Sciences, Journal Year: 2025, Volume and Issue: 8(1), P. 58 - 75

Published: March 27, 2025

Ear diseases are characterized by a range of symptoms, including balance disturbances, delayed speech development in children, headaches, fever, and hearing loss. In order to prevent further complications, it is essential that these conditions diagnosed treated timely manner. The traditional primary diagnostic method has been otoscope examination otolaryngologists. However, the accuracy this approach contingent upon clinician's expertise quality equipment used, which can render susceptible misdiagnosis. Incorrect diagnoses may result administration antibiotics unnecessarily, disease progression, other adverse consequences. This study aims evaluate efficacy computationally efficient machine learning models classifying ear images. To enhance classification accuracy, Histogram Oriented Gradients (HOG) was employed for feature extraction, optimization algorithms were utilized selection. Whale Optimization Algorithm (WHO) observed exhibit notable selection informative features k-Nearest Neighbors (kNN) model, achieving 92.6%. Furthermore, Support Vector Machine (SVM) model achieved an 92% using map comprising selected algorithms. experimental findings emphasize potential strategic enhancing performance classical classification. By employing techniques such as HOG algorithms, attain accuracies on par with those more resource-intensive deep approaches. Such developments facilitate creation accessible tools, particularly beneficial resource-constrained clinical settings. provide basis research aimed at precision learning-based medical imaging.

Language: Английский

Enhancing intra-aural disease classification with attention-based deep learning models DOI Creative Commons
Furkancan Demircan, Murat Ekіncі, Zafer Cömert

et al.

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 21, 2025

Language: Английский

Citations

1
Enhanced Classification of Ear Disease Images Using Metaheuristic Feature Selection DOI Creative Commons
Furkancan Demircan, Murat Ekіncі, Zafer Cömert

et al.

Sakarya University Journal of Computer and Information Sciences, Journal Year: 2025, Volume and Issue: 8(1), P. 58 - 75

Published: March 27, 2025

Ear diseases are characterized by a range of symptoms, including balance disturbances, delayed speech development in children, headaches, fever, and hearing loss. In order to prevent further complications, it is essential that these conditions diagnosed treated timely manner. The traditional primary diagnostic method has been otoscope examination otolaryngologists. However, the accuracy this approach contingent upon clinician's expertise quality equipment used, which can render susceptible misdiagnosis. Incorrect diagnoses may result administration antibiotics unnecessarily, disease progression, other adverse consequences. This study aims evaluate efficacy computationally efficient machine learning models classifying ear images. To enhance classification accuracy, Histogram Oriented Gradients (HOG) was employed for feature extraction, optimization algorithms were utilized selection. Whale Optimization Algorithm (WHO) observed exhibit notable selection informative features k-Nearest Neighbors (kNN) model, achieving 92.6%. Furthermore, Support Vector Machine (SVM) model achieved an 92% using map comprising selected algorithms. experimental findings emphasize potential strategic enhancing performance classical classification. By employing techniques such as HOG algorithms, attain accuracies on par with those more resource-intensive deep approaches. Such developments facilitate creation accessible tools, particularly beneficial resource-constrained clinical settings. provide basis research aimed at precision learning-based medical imaging.

Language: Английский

Citations

0