T-CAIREM Trainee Rounds: Brokoslaw Laschowski and Michael Lee
DATE: May 30, 2022 (Monday)
TIME: 12pm to 1pm
PRESENTERS: Brokoslaw Laschowski and Michael Lee
Postdoctoral research fellow, Temerty Faculty of Medicine and the Toronto Rehabilitation Institute
Development of wearable computer vision systems powered by deep learning for automated control and decision making of robotic leg prostheses and exoskeletons.
Robotic leg prostheses and exoskeletons can replace the propulsive function of impaired or amputated biological muscles and allow users with mobility impairments to perform daily locomotor activities that require power generation. However, the current locomotion mode recognition systems being developed for automated control and decision making rely on mechanical, inertial, and/or neuromuscular sensors, which inherently have limited prediction horizons (i.e., analogous to walking blindfolded). In this talk, Dr. Laschowski will present his cutting-edge research on the development of wearable computer vision systems powered by deep learning to predict the oncoming walking environment prior to physical interactions, therein allowing for more accurate and robust control decisions. These state-of-the-art environment recognition systems serve to improve the automated control and decision-making of next-generation robotic leg prostheses and exoskeletons for daily locomotor assistance and rehabilitation.
Dr. Brokoslaw Laschowski is a postdoctoral research fellow in the Temerty Faculty of Medicine at the University of Toronto and the Toronto Rehabilitation Institute. He specializes in using mathematical, computational, and machine learning methods to optimize the design and control of humans interacting with wearable robotic systems and technologies. Applications of his research include rehabilitation robotics, neural engineering, human-computer interaction, and wearable assistive devices (i.e., exoskeletons and robotic leg prostheses). His clinical research focuses on assisting individuals with mobility impairments due to aging and/or physical disabilities such as stroke, cerebral palsy, osteoarthritis, Parkinson’s disease, amputation, and spinal cord injury.
Michael Kyung Ik Lee
M.Sc. student, Temerty Faculty of Medicine, Laboratory Medicine & Pathobiology
Developing a Compound Computational Pathology Workflow for Automated and Integrated Analyses of Morphologic and Molecular Histopathological Features
Despite recent innovations in deep learning, attempts to automate the integration of histomorphologic and molecular information found on respective H&E- and IHC-stained tissue sections are scarce. In this presentation, I will discuss how I have developed a fully automated workflow that incorporates both H&E-stained sections and their accompanying IHC studies using computer vision tools, such as deep learning and scale-invariant feature transform (SIFT). I will also be discussing the performance of the workflow on analysis and subclassification tasks of diffuse glioma cases as a proof-of-concept.
Mr. Lee holds an Honours Bachelor of Science from the University of Toronto with a Specialization in Pathobiology, Major in Biochemistry, and a Minor in Latin. He recently defended his thesis for Master of Science in Laboratory Medicine & Pathobiology. During his undergraduate studies, Lee was involved with basic science research on epigenetic pathways of endothelial cells, completing an undergraduate thesis in Dr. Philip Marsden’s laboratory. For his master’s, he branched out to learn more about digital pathology, a dynamic field that can revolutionize diagnostic pathology, and in turn contribute to the development of various computational tools to move the field forward.