(Nov 20) Deep Learning Approaches for Biomedical Image Analysis

In recent years, Deep Learning (DL) application has shown promising results in conducting AI tasks, such as computer vision and speech recognition. Convolutional Neural Network (CNN) models have been utilized for identifying special objects targeted as preventive, detective, and diagnostic means in predictive medicine. However, automatic segmentation and classification of biomedical images have been a challenging task, especially for Magnetic Resonance Images (MRI) applied to diagnose and treatment of brain diseases. Moreover, it is a time-consuming procedure that requires trained biomedical experts to manually segment or annotate such MRI datasets. This talk will introduce Deep Learning (DL) approaches (Mask-RCNN and U-Net) which are the state-of-the-art convolutional neural network algorithms for object analysis in the biomedical domains.  Mask-RCNN and U-Net have been applied to detection and segmentation of biomedical images obtained from specialized pathologic areas, such as Brain or Oral Diseases. In this presentation, we will demonstrate that DL based Mask-RCNN can be used to perform highly efficient automatic segmentation and classification of pathology imaging informatics.