Pothina Praveena

Publications

• Research Article   
  MCRIP-UNet: A modified U-Net architecture for semantic segmentation of images for breast cancer detection
  Author(s): Pothina Praveena* and N. Suresh Kumar
  
  We detail a deep learning architecture that performs an effective nuclei segmentation and recognition without much intervention with the system from images stained by MCRIP-UNet for breast cancer. Based on deep learning, MCRIP UNet is a multi-dimensional cross-reconstruction inverted pyramid network introduced for enhancing precision as well as acceleration in the field of medical image segmentation. Three different encoder architectures are applied to extract feature information from each modality by using multimodal magnetic resonance images as input. At the same resolution level, the retrieved feature data is first fused. A dual band cross reconstruction attention module is used for multimodal feature fusion and refinement. In order to achieve the goal of breast tissue segmentation, a pyramidal-shaped decoder is used at each stage of the decoder to integrate the features of differe.. Read More»
  

  Abstract PDF