CISE Seminar: November 1, 2019 – Konstantinos N. Plataniotis, University of Toronto
BU Photonics Building
8 St. Mary’s Street, PHO 203
3:00pm-4:00pm
Konstantinos N. Plataniotis
University of Toronto
Image Processing and Machine Learning for Histopathology and Radiomics
Due to the proliferation of whole-slide-imaging (WSI) digital scanners it is now possible to leverage color image processing, analysis, and machine learning techniques, such as deep learning to process the digital pathology images in hopes to derive, diagnosis and prognosis markers. The convergence of digital image processing and pathology gave rise to a new research area known as computational pathology. Computational pathology greatly enhances diagnostic accuracy and allows a variety of pathology tasks to be completed with greater efficiency. This presentation will offer a general introduction to computational pathology, it will outline and discuss image processing tasks needed for the successful implementation of a computational pathology pipeline, and it will offer overview and insights on how data-driven solutions such as deep neural networks can be used to derive markers from digital pathology slides.
The lecture will demonstrate how image pre-processing can boost generalizability of pre-trained computational pathology solutions. An elegant normalization solution to combat variability in color and noise levels among cross-institution pathology images will be discussed. The lecture will introduce an unsupervised solution to decompose stain mixing in pathology images. In this way, start-of-the-art grayscale image analysis techniques can be readily applied in computational pathology tasks. It will also be shown that a diagnostic system combining deep learning and prior histological knowledge can provide useful diagnostic/prognostic markers. For the important invasive breast cancer diagnosis it will be shown that detection of histological abnormalities can be simplified to identification of common patterns in normal breast images.
Lastly, open research issues such as strategies to develop generalizable solutions in computational pathology with limited amount of training data, how to combine computational pathology and radiomics markers, and future trends will be briefly discussed.
Konstantinos N. Plataniotis, Bell Canada Chair in Multimedia, is a Professor with the ECE Department at the University of Toronto. His current research interests are: machine learning, adaptive systems & pattern recognition, image & signal processing, communications systems, and big data analytics. He is a registered professional engineer in Ontario, Fellow of the IEEE and Fellow of the Engineering Institute of Canada. Dr. Plataniotis was the IEEE Signal Processing Society inaugural Vice President for Membership (2014-2016) and the General Co-Chair for the IEEE GlobalSIP 2017 (November 2017, Montreal, Q.C.). He co-chairs the 2018 IEEE International Conference on Image Processing (ICIP 2018), October 7-10, 2018, Athens Greece, and the 2021 IEEE International Conference in Acoustics, Speech and Signal Processing (ICASSP 2021), Toronto, ON, Canada.
Faculty Host: Janusz Konrad
Student Host: Arian Houshmand