Sparse Deep Learning for Cell Type Classification in Breast Cancer Using Multiplex Imaging

Imaging Mass Cytometry (IMC) is an important imaging technology for breast cancer research. However, there is still a lack of research on the development of efficient deep learning algorithms for classification of cell types based on IMC data. In this study, we propose a sparse deep learning approach and assess the importance of antibody markers that contribute to cell classification. The results of the experiment show that the key markers we identified align closely with those traditionally used for manual cell classification, which provide tremendous potential for future biological research.