Dr. Ahmad P. Tafti will give the sixth talk in The USM Data Science Ensemble, a seminar series focused on the intersection of data science and real-world applications. We invite you to join us for this in-depth look at a practical application of data science in the real world, by joining this moderated Zoom link.
Dr. Tafti's talk, "Multi-Class Semantic Segmentation of Medical Images using Deep Learning Computational Methods," will look at the serious role that medical image segmentation plays in a variety of healthcare applications, by automatically partitioning anatomical structures into multiple regions of interest so that they can be examined separately. In the last few years, deep learning has advanced as artificial intelligence algorithms have been increasingly used to interpret and discover patterns from medical images, and they have already demonstrated promising results in image segmentation and object detection. In this one-hour tutorial, he will present, from the computational side, state-of-the-art deep learning-powered image segmentation on knee x-ray images. We will work on a deep learning model, called U-Net, and practically explore what it is, what it does, and how.
Ahmad P. Tafti is an Assistant Professor in the Computer Science Department at University of Southern Maine, with a deep passion for improving health informatics using multi-modal medical data combined with advanced computational methods. Dr. Tafti’s major interests are AI, machine learning, and computational health informatics. He completed his PhD in Computer Science at the University of Wisconsin-Milwaukee, where some parts of his international studies were carried out at Oracle Education Center, Malaysia, Vienna University of Technology, Austria. He won the General Electric Honorable Mention Award. Dr. Tafti has published over 45 peer-reviewed papers in prestigious journals and conferences (e.g., CVPR, AMIA, ISVC, JMIR, PLOS, IEEE Big Data), addressing medical text and medical image analysis and understanding using advanced computational strategies. In addition, Dr. Tafti has served as a workshop organizer, steering committee member, technical reviewer, and a program committee member for several reputable conferences and journals, including KDD 2017, AMIA, IEEE ICHI, ISMCO, ISVC, IEEE Journal of Biomedical and Health Informatics, and International Journal of Computer Vision and Image Processing. He was awarded a NVIDIA GPU Grant for his accomplishments in deep learning community.