Diffuse Large B-Cell Lymphoma stands as the most prevalent form of non-Hodgkin lymphoma worldwide, constituting approximately 30 percent of cases within this diverse group of blood cancers affecting the lymphatic system. This study addresses the challenges associated with the accurate DLBCL segmentation and classification, including difficulties in identifying and diagnosing DLBCL, manpower shortage, and limitations of manual imaging methods. The study highlights the potential of deep learning to effectively segment and classify DLBCL types. The implementation of such technology has the potential to extract and preprocess image patches, identify, and segment the nuclei in DLBCL images, and classify DLBCL severity based on segmented nuclei counting.
