Deep Learning Based Classification for AFM Images of Cancer Cells

Deep learning has exhibited promising outcomes particularly in computer vision. In the medical field, deep learning has been successfully utilized to improve diagnostic accuracy and efficiency in cancer diagnosis. Given the complexity of cancer, early detection and diagnosis remain challenging tasks. Atomic Force Microscopy (AFM) has been reported as an important tool for the diagnosis of both benign and malignant cells, due to its ability to provide information on cell mechanics and ultrastructures. Therefore, it is meaningful to use deep learning algorithms to classify cancer cells based on AFM images. This classification can include identifying cancer subtypes and differentiatoins between benign and malignant properties. The purpose of this work is to review the progress and research implications in the area of deep learning based classification for AFM images of cancer cells.