Comprehensive Study on Lymphoma Detection Using Deep Learning
Naveen Kumar H N, H S Bhanu, Muhammed Junaid, S Supritha, S. P. Ramya, Khandekar Srushti · 2025
Lymphoma represents a form of cancer, affecting the lymphatic system together with multiple variants requiring precise diagnosis for suitable therapy. According to the pathological manifestations, lymphoma is divided into Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). NHL includes Mantle Cell Lymphoma (MCL), follicular lymphoma (FL), and Chronic Lymphocytic Leukemia (CLL). The application of artificial intelligence (AI). Histopathology slides, a cornerstone of lymphoma diagnosis, exhibit significant spatial heterogeneity within and between samples. Capturing and leveraging these complex spatial patterns and cellular interactions using deep learning models requires sophisticated architectures beyond simple convolutional neural networks (CNNs). This survey provides a comprehensive study of existing Deep Learning methods which are used to detect and classify lymphoma subtypes in histopathological images. The benchmark datasets for the lymphoma subtypes are discussed in a coherent manner. This survey presents a comprehensive summary on state-of-the-art methods for lymphoma subtype classification. This review could help in selecting the best framework by providing performance metrics, data dependencies, scopes in generalizability and boundaries of models to identify and diagnose different lymphoma types. The study also examines complex research gaps by presenting detailed analysis of open-ended research opportunities in lymphoma detection.