RDSC Scientists to Speak at URMC’s Multiphoton and Analytical Imaging Center

Machine and Deep Learning for Improved Medical Image Analysis

The Rochester Data Science Consortium will speak at the University of Rochester Medical Center as part of its ongoing 2020 Spring Imaging Seminar Series, co-hosted by the Multiphoton and Analytical Imaging Center (MAGIC) and the Center for Advanced Light Microscopy and Nanoscopy​ (CALMN).

Steven Smith, (Vice President), Edgar Bernal, Ph.D., (Associate Director and Senior Research Scientist), Wencheng Wu, Ph.D., (Senior Research Scientist), and Beilei Xu, Ph.D., (Senior Research Scientist) will present on the topic of “Machine and Deep Learning for Improved Medical Image Analysis,” providing a high-level overview of machine learning and deep learning for image analysis, as well as a cursory discussion of example applications to the medical and healthcare fields.

The event will take place on Tuesday, March 10th, from 12:00pm-1:00pm, at Strong Memorial Hospital (601 Elmwood Avenue) in the Ryan Case Method Room (1-9576). A hospital map can be found here.

More information about the 2020 Imaging Seminar Series, along with upcoming speakers, can be found here. We hope to see you there.

A full abstract of the RDSC’s talk can be found below:

Abstract:

With the advent of machine learning and deep learning, machines can now achieve and sometimes exceed human-level performance in many tasks.  This has generated a great deal of interest in the research community, particularly in the field of computer vision.  Typical learning-based vision tasks include image classification, object detection, image segmentation, and scene understanding.  The success of existing frameworks in these areas can be leveraged to improve medical image analysis through better data-based reasoning and decision-making, and by automating labor-intensive or error-prone manual steps.

In this talk, we will provide a high-level overview of machine learning and deep learning for image analysis, as well as a cursory discussion of example applications to the medical and healthcare fields.  We will focus our discussion on two of the most common vision tasks, namely image classification and semantic segmentation.  We will start by describing the typical workflow for each task, introduce a few commonly used neutral network architectures, and conclude with practical examples.