ReliefF-ResNet-101: An Efficient Deep Learning Framework for Oral Squamous Cell Carcinoma Detection
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Oral squamous cell carcinoma (OSCC) is a leading cause of cancer-related deaths globally, with 177,757 fatalities reported in 2020 World Health Organization. Early detection significantly improves survival rates. This study proposes an automated machine-learning approach for OSCC detection using histopathological images. Deep features extracted from the pre-trained model (ResNet-101) achieved 91.61\% classification accuracy. Furthermore, using ReliefF-based feature selection filters enhanced performance by 97.3\%. The ReliefF-ResNet-101 deep feature-trained support vector machine model required only 500 features. The comparative analysis confirmed the reliability and superiority of the proposed ReliefF-ResNet-101 framework over existing feature selection methods and state-of-the-art approaches. These findings suggest potential clinical applications for aiding OSCC diagnosis.
Abstract
Oral squamous cell carcinoma (OSCC) is a leading cause of cancer-related deaths globally, with 177,757 fatalities reported in 2020 World Health Organization. Early detection significantly improves survival rates. This study proposes an automated machine-learning approach for OSCC detection using histopathological images. Deep features extracted from the pre-trained model (ResNet-101) achieved 91.61\% classification accuracy. Furthermore, using ReliefF-based feature selection filters enhanced performance by 97.3\%. The ReliefF-ResNet-101 deep feature-trained support vector machine model required only 500 features. The comparative analysis confirmed the reliability and superiority of the proposed ReliefF-ResNet-101 framework over existing feature selection methods and state-of-the-art approaches. These findings suggest potential clinical applications for aiding OSCC diagnosis.