A Deep CNN-BiGRU model for Lymphoma and Residual Masses Recognition

Medical science has been revolutionised by advances in deep learning, particularly in the detection and treatment of disease, particularly in the field of oncology. Among the most worrying cancers are those that affect the lymphatic system, known as lymphomas. Due to their increasing prevalence, early detection of lymphomas has become critical to improving clinical outcomes and patient survival. Advances in medical imaging andimage analysis have enabled the development of sophisticated deep learning models capable of automatically detecting and characterising lymphatic lesions on radiological images. In this context, we proposed an innovative approach “CNN-BiGRU model” based on ResNet50, a deep convolutional neural network recognised for its ability to perform exceptionally well in image recognition challenges, and bidirectional recurrent networks (BiGRU) for automatic lymphoma recognition from multimodal medical images. By integrating the capabilities of these deep learning models, our method aims to provide valuable assistance to clinicians for early and precise identification of lymphoma, enabling faster and better therapeutic intervention.