Deep Learning Approaches For Lymphoma Prediction: Leveraging VGG16 and DenseNet Architectures

Blood Cancer is a gruesome disease that causes the abnormal growth of cells in the blood, bone marrow, and lymphatic system. It uses various CT scans for the recovery or detection of blood cancer. This paper aims to predict blood cancer at an early rate, by using different Deep Learning (DL) technologies like VGG16 and DenseNet. In the medical sciences or healthcare field DL algorithms work efficiently, with innovative ideas. With their different layers, DL algorithms work better on the data set of blood cancer, the data set has mainly three classes named lymph_cll, lymph_fl, and lymph_mcl. DenseNet achieved greater accuracy than the VGG16 model. The DenseNet model achieves high performance because each layer is densely interconnected with every adjacent layer, this leads to effective information transfer by feature reuse and enhanced gradient flow, leading to its excellent results in tasks involving the classification of blood cancer, through the utilization of these developments in DL, this study seeks to progress early detection techniques and ultimately enhance patient care results in the treatment of blood cancer.