Breast cancer diagnosis through knowledge distillation of Swin transformer-based teacher-student models

Article Oncology

Cancer statistics, 2022

Rebecca L. Siegel et al.

Summary: Each year, the American Cancer Society compiles data on cancer occurrence and outcomes in the United States. The latest data shows that breast and prostate cancer progress has stagnated, while lung cancer has shown improvements in survival rates. Lung cancer incidence for advanced disease has declined while localized-stage rates have increased, resulting in higher survival rates. Mortality patterns align with incidence trends, with lung cancer deaths declining rapidly, breast cancer deaths slowing, and prostate cancer deaths stabilizing.

CA-A CANCER JOURNAL FOR CLINICIANS (2022)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Swin Transformer V2: Scaling Up Capacity and Resolution

Ze Liu et al.

Summary: This paper presents techniques for scaling Swin Transformer up to 3 billion parameters and the ability to train with high-resolution images. By increasing the capacity and resolution, Swin Transformer achieves new records on four representative vision benchmarks. Several novel technologies are proposed to address training instability and effectively transfer models from low-resolution to high-resolution. Using these techniques and self-supervised pre-training, a strong 3 billion Swin Transformer model is successfully trained, achieving state-of-the-art accuracy on various benchmarks.

2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) (2022)

Add to Collection

Article Engineering, Biomedical

A new transfer learning based approach to magnification dependent and independent classification of breast cancer in histopathological images

Said Boumaraf et al.

Summary: This paper proposes a new transfer learning-based approach to automated classification of breast cancer from histopathological images using ResNet-18. The method refines the deep neural network on histopathological images through block-wise fine-tuning strategy, achieving promising results on the BreaKHis dataset.

BIOMEDICAL SIGNAL PROCESSING AND CONTROL (2021)

Add to Collection

Article Computer Science, Information Systems