Recent Seminars

Rethinking Convolutional Neural Networks (CNNs)


Dr. C.-C. Jay Kuo is a University of Southern California Distinguish Professor and Directory of its Media Communications Laboratory. Dr. Kuo received his Ph.D. from MIT in 1987. He has served as editor for 12 international journals and co-authored around 250 journal papers, 900 conference papers and 14 books. He was the guest speaker at a lecture organized by the MGH Gordon Center. Below is the presentation summary provided through the courtesy of Dr. Kuo.

Dr. Kuo delivering his presentation at the mgH Gordon Center

The superior performance of Convolutional Neural Networks (CNNs) has been demonstrated in many applications such as image classification, detection and processing. Yet, the CNN solution has its own weaknesses such as robustness against perturbation, scalability against the class number and portability among different datasets. Furthermore, CNN’s working principle remains a mystery. In this talk, Dr. Kuo first explained the reasons behind the superior performance of CNNs. Then, he presented an alternative solution, which is motivated by CNNs yet allows rigorous and transparent mathematical treatment, based on a data-driven Saak (Subspace approximation with augmented kernels) transform. The kernels of the Saak transform are derived from the second-order statistics of inputs in a one-pass feedforward way. Neither data labels nor backpropagation is needed in kernel determination. The pros and cons of CNNs and multi-stage Saak transforms were compared.

Data Analytics in Operations Management


Dr. Oleg S. Pianykh is the Director of Medical Analytics at the Department of Imaging, Massachusetts General Hospital, and Assistant Professor at Harvard Medical School. With academic background in applied computer science, He has been actively working in the field of innovative healthcare for the past 20 years. His scientific work ranged from research on digital imaging and data-driven clinical workflow to publishing and teaching advanced graduate courses at Harvard and other leading universities. On the applied side, Dr. Pianykh has served as a CIO for a state-wide healthcare network, and participated in many consulting/startup initiatives. His current interests include bid data analysis, operations management, and information technology in healthcare.
Dr. Pianykh was the guest speaker at a lecture organized by the MGH Gordon Center. Below is his presentation’s summary.

With medical technology becoming increasingly complex, data volumes – increasingly high, and expected outcomes – more demanding, the cost of medical errors, processing delays and guesswork becomes prohibitively high. To deal with these challenges, contemporary radiology has to learn how to use its data to produce optimal decisions and operational strategies. Transforming radiology data into the most effective and objective problem solver was the main idea behind the Medical Analytics Group (MAG) project, launched by the Department of Radiology at Massachusetts General Hospital. The principal purpose of MAG is to apply data science to routine problems, looking for the best possible solutions. Their current projects include identifying hidden operational bottlenecks, building predictive workflow models, developing optimal scheduling strategies, analyzing utilization and productivity limits, studying processing quality and satisfaction, and many more: MAG work is driven by current needs, not limited to any particular domain. All MAG projects have to be implemented in real clinical environments; all have to be verified and proven to work with objective data analysis. In this presentation Dr. Pianykh shared the Medical Analytics Group's most interesting results, important successes, and thought-provoking challenges.

Dr. Pianykh delivering his presentation at the MGH Gordon Center

Novel Fiber-Based Optical Systems for Biomedical Applications


Dr. Rayan Zaman is an Assistant Professor in the Department of Biomedical Engineering at the Virginia Commonwealth University. She has developed a Circumferential-Intravascular-Radioluminescence-Photoacoustic-Imaging (CIRPI) system to detect and characterize vulnerable plaques in human, mice and porcine models. She was the guest speaker at a lecture organized by the MGH Gordon Center.

Optical imaging offers many exciting opportunities to develop minimally invasive, low-cost solutions for the detection and treatment of diverse diseases in cardiovascular and many other areas that are yet to be explored. Dr. Rayan Zaman discussed four major clinical applications of optical systems:
Measurement of changes in blood flow of skin blood vessels using Speckle Contrast Imaging
Pharmacokinetics analysis of drug during treatment of ophthalmic diseases using Fluorescence Spectroscopy System;
Diagnosis of disease in cardiology using Balloon-Enabled Fluorescence/Radioluminescence Optical Imaging System;
Detection/characterization of atherosclerotic plaque using Circumferential-Intravascular-Radioluminescence-Photoacoustic-Imaging (CIRPI) System.

Dr. Rayan Zaman delivering her presentation at the MGH Gordon Center

Seeing the Unseen in Patients: Advancing Disease Prevention and Treatment through Microimaging


Dr. Guillermo Tearney is Professor of Pathology at Harvard Medical School, Mike and Sue Hazard Family MGH Research Scholar, an Affiliated Faculty member of the Harvard-MIT Division of Health Sciences and Technology (HST), Fellow of the American College of Cardiologists (FACC), Fellow of the College of American Pathologists (FCAP) and heads his own lab at the Wellman Center for Photomedicine at the Massachusetts General Hospital. Dr. Tearney received his MD magna cum laude from Harvard Medical School and received his PhD in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology. He was the guest speaker at a lecture organized by the MGH Gordon Center. Below is the presentation summary provided through the courtesy of Dr. Tearney.

Today's gold standard for medical diagnosis is histology of excised biopsies or surgical specimens where tissue is taken out of the body, processed, sectioned, stained and looked at under a light microscope by a pathologist. There are many limitations of this technique, including the fact that it is inherently invasive, time consuming, costly, and dangerous for some organs. Furthermore, oftentimes the diseased tissue is not readily seen by visual inspection and as a result the tissue is sampled at a random location, which can be highly inaccurate.

If we could instead conduct microscopy inside the body, then we could provide tools for screening, targeting biopsies, making primary disease diagnosis, and guiding intervention on the cellular basis. This promise of has motivated the development of a new field, termed in vivo microscopy, the goal of which is to obtain microscopic images from living human patients. Two in vivo microscopy technologies, confocal microscopy and optical coherence tomography, are currently available and in clinical use. Upcoming developments, including whole organ microscopy, swallowable microscopy capsules, molecular imaging, and very high resolution microscopic devices are in the pipeline and will likely revolutionize how disease is diagnosed and how medicine is practiced in the future. Both of these patch-based techniques capture local image features through particular learning processes. They have been demonstrated to improve conventional MAP methods in reconstructing simulated and patient PET data, showing their potential for clinical applications.

Dr. Guillermo Tearney delivering his presentation at the MGH Gordon Center

Seminar: Advancing PET Image Reconstruction with Machine Learning


Dr. Jing Tang received her PhD degree in Electrical Engineering from the University of Illinois at Urbana-Champaign and had her postdoctoral training in Radiology at the John Hopkins School of Medicine. She was employed as a principle imaging physicist at the Philips Healthcare before joining the Department of Electrical and Computer Engineering at Oakland University, where she currently holds an appointment as an associate professor. She has received a number of federal awards to support her research on the development and application of PET/MR imaging techniques. Dr. Tang was the guest speaker at a lecture organized by the MGH Gordon Center. Below is her presentation’s summary.

Machine learning has shown its promises to empower medical imaging, mainly in image analysis. The talk presented Dr. Tang’s investigation of integrating machine learning techniques into the other major component of medical imaging, image reconstruction. Dr. Tang proposes to advance maximum a posteriori (MAP) PET image reconstruction using two learning based methods.

The first one is through building a dictionary learning (DL) based prior and the second by designing an artificial neural network (ANN) to enhance smoothness MAP reconstructed images from different regularizing weights. The DL-based MAP reconstruction technique forms its regularization using the sparse representation of PET images based on dictionaries trained from various images. The ANN enhancement scheme solves the regression problem of mapping between multiple reconstructed image versions and an enhanced image through learning from examples.

According to Dr. Tang, both techniques have a potential for clinical applications due to their capacity to improve conventional MAP methods in reconstructing simulated and patient PET data.

Dr. Jing Tang delivering his presentation at the MGH Gordon Center

From Academia to Industry and Back Again: Transitioning Technology & Changing Employers


Dr. Lee Josephson is a chemical biologist and an inventor on more than 20 patents in the fields of medical imaging. He has been an NIH grant Principal Investigator and a Vice-President and Chief Scientific Officer of a diagnostic drug pharmaceutical company. He was the guest speaker at a lecture organized by the MGH Gordon Center.

Dr. Josephson's discussed the transition from academic to industrial employers, including some of the different types of employers in the Boston area. His presentation also covered the difference between what companies and academic institutions look for in new hires. Dr. Lee used his personal case examples of success and failure to describe issues that arise during the transition from academic science to industry.

Dr. Lee Josephson delivering his presentation at the MGH Gordon Center

Seminar: Collaborative Innovation at Partners HealthCare


Dr. Glenn Miller is Innovation’s Market Sector Leader for Radiology, Anesthesiology, Neurology, Neurosurgery and Psychiatry at Partners Healthcare. He has nearly 30 years of experience in the clinical laboratory industry and personalized medicine.

The role of Partners Healthcare (PHC) Innovation is to assist investigators in the sustainable support for their work, intellectual property protection of their inventions and commercialization of their discoveries. Dr. Miller's lecture discussed how PHC Innovation can help investigators protect and commercialize their scientific inventions through different mechanisms such as patents, licenses, industry collaboration and venture investments.

Dr. Glenn Miller delivering his presentation at the MGH Gordon Center

Seminar: A Projected Filter Algorithm for Dynamic SPECT


Dr. Youssef Qranfal has served as Professor of Applied Mathematics at Wentworth Institute of Technology in Boston, Massachusetts since September 2015. His recent research focuses on optimization, operation research, statistics, and their applications. Prior to starting his career at WIT, Dr. Qranfal has worked in industry as an engineer in applied mathematics and computer science. He has authored many technical papers on applied mathematics to various fields such as medical imaging. They have been published in peer-reviewed journals, presented at technical conferences, and appeared in the proceedings of those conferences.

Images and visualization have become increasingly important in many areas of science and technology. Advances in hardware and software have allowed computerized image processing to become a standard tool in many scientific applications, including medical imaging. In this talk, Dr. Qranfal demonstrated how he models and solves the inverse problem of reconstructing a dynamic medical image where the signal strength changes substantially over the time required for data acquisition. His group uses a stochastic approach based on a Markov process to model the problem. Dr. Qranfal and his collaborators introduced a novel proximal approach and applied it during the Kalman filter algorithm to ensure positivity and spatial regularization. They have tested their method for the case of image reconstruction in time-dependent single photon emission computed tomography (SPECT). According to Dr. Qranfal, numerical results corroborate the effectiveness of his approach.

Prof. Qranfal discusses reconstruction of time-varying SPECT images