Enhancing Oral Squamous Cell Carcinoma Detection Using Histopathological Images: A Deep Feature Fusion and Improved Haris Hawks Optimization-Based Framework DOI Creative Commons
Amad Zafar,

Majdi Khalid,

Majed Farrash

et al.

Bioengineering, Journal Year: 2024, Volume and Issue: 11(9), P. 913 - 913

Published: Sept. 12, 2024

Oral cancer, also known as oral squamous cell carcinoma (OSCC), is one of the most prevalent types cancer and caused 177,757 deaths worldwide in 2020, reported by World Health Organization. Early detection identification OSCC are highly correlated with survival rates. Therefore, this study presents an automatic image-processing-based machine learning approach for detection. Histopathological images were used to compute deep features using various pretrained models. Based on classification performance, best (ResNet-101 EfficientNet-b0) merged canonical correlation feature fusion approach, resulting enhanced performance. Additionally, binary-improved Haris Hawks optimization (b-IHHO) algorithm was eliminate redundant further enhance leading a high rate 97.78% OSCC. The b-IHHO trained k-nearest neighbors model average vector size only 899. A comparison other wrapper-based selection approaches showed that results statistically more stable, reliable, significant (

Language: Английский

Enhancing Oral Squamous Cell Carcinoma Detection Using Histopathological Images: A Deep Feature Fusion and Improved Haris Hawks Optimization-Based Framework DOI Creative Commons
Amad Zafar,

Majdi Khalid,

Majed Farrash

et al.

Bioengineering, Journal Year: 2024, Volume and Issue: 11(9), P. 913 - 913

Published: Sept. 12, 2024

Oral cancer, also known as oral squamous cell carcinoma (OSCC), is one of the most prevalent types cancer and caused 177,757 deaths worldwide in 2020, reported by World Health Organization. Early detection identification OSCC are highly correlated with survival rates. Therefore, this study presents an automatic image-processing-based machine learning approach for detection. Histopathological images were used to compute deep features using various pretrained models. Based on classification performance, best (ResNet-101 EfficientNet-b0) merged canonical correlation feature fusion approach, resulting enhanced performance. Additionally, binary-improved Haris Hawks optimization (b-IHHO) algorithm was eliminate redundant further enhance leading a high rate 97.78% OSCC. The b-IHHO trained k-nearest neighbors model average vector size only 899. A comparison other wrapper-based selection approaches showed that results statistically more stable, reliable, significant (

Language: Английский

Citations

0