Saikat Roy

Saikat Roy is a visiting researcher at the Diagnostic Image Analysis Group (DIAG), where he works on advanced deep learning methods for medical image analysis. His research focuses on large-scale supervised representation learning and 3D medical image segmentation, with an emphasis on developing strong, generalizable backbone architectures. He is a PhD Student at the German Cancer Research Center(DKFZ), Heidelberg, Germany and Heidelberg University, Germany. He has led the development of Transformer and ConvNeXt architectures that have demonstrated strong performance across diverse CT and MR segmentation benchmarks. He was previously a Masters Student at University of Bonn, Germany and was associated with the German Center for Neurodegenerative Diseases (DZNE), Germany.

Member of:

• Diagnostic Image Analysis Group