IEEE Toronto Section

IEEE

Archive for the ‘Signals & Computational Intelligence’ Category

A framework for general purpose digital pathology image analysis, using machine learning methods to identify cancer subsets and immunotherapy biomarkers

Friday, July 7th, 2017

Monday July 17, 2017 at 4:00 p.m. Dr. Trevor McKee, STTARR Innovation Research Centre for Cancer Research, will be presenting “A framework for general purpose digital pathology image analysis, using machine learning methods to identify cancer subsets and immunotherapy biomarkers”.

Day & Time: Monday July 17, 2017
4:00 p.m. – 5:00 p.m.

Speaker: Dr. Trevor McKee
STTARR – Innovation Research Centre for Cancer Research
Toronto, Ontario, Canada

Location: Room ENG101
George Vari Engineering Building (intersection of Church & Gould)
Ryerson University
245 Church St, Toronto, M5B 1Z4

Contact: Alireza Sadeghian, Alex Dela Cruz

Organizers: Signals & Computational Intelligence Chapter

Abstract: Histological staining, interpreted by a pathologist, has remained the gold standard for cancer diagnosis and staging for over 100 years. There is a growing need for better – and more personalized – cancer treatments, to provide oncologists with the tools they need to best treat their patients. The advent of “molecular medicine”, or targeted therapeutic strategies that rely on knowledge of particular mutations in a cancer in order to tailor treatment, has improved cancer therapy for many patients. This has led to the use of companion diagnostics, in which tumor biopsies are stained for a specific marker or set of markers, using immunohistochemical approaches. The information obtained from the degree of staining or spatial arrangement of stained cells within the tumor helps to identify tumor molecular subclasses that may benefit from such tailored therapeutic approaches.

The increase in the number of slides being stained for specific markers and used in diagnosis, along with the increased need for quantitative assessment of the degree of staining, number of cells, or spatial arrangement of cells within the tumor, has increased the volume and type of work that pathologists encounter in their diagnostic workflow. Our team works on the development of tools for quantitative digital pathology analysis that can benefit pathologists, by building and validating semi-automated algorithms for cellular quantification and intensity scoring of stained slides. We use machine learning methods to learn features that distinguish different morphological regions from pathologist annotations. These are then fed into a tissue segmentation and classification framework to break the tissue down into its components, either on the individual cell level, or the glandular level. Staining intensity is quantified following colour deconvolution of the individual stain components, and reporting metrics are designed, in close collaboration with pathologists and biological scientists, to identify the appropriate outputs for comparing between treatment groups or different cancer types.

The use of multiplexed digital pathology stains allows us to build a generalized analytical framework to perform “tissue cytometry”. This new technology can extract quantitative image-derived features in a reproducible and robust fashion, providing clinicians and biological scientists with tools to measure previously inaccessible phenomena, like measuring the hypoxic gradient directly within tumor sections, or comparing glucose uptake to lactic acid production in the same tumor sample. This approach establish the foundation for a bridge between traditional morphometric assessment of tumor biopsies, and the detailed spatially resolved chemical and molecular content maps of each tumor, providing an invaluable toolkit for the discovery of cancer molecular subtypes, and development of therapeutic interventions.

Biography: Dr. Trevor McKee received his Ph.D. in Biological Engineering from the Massachusetts Institute of Technology in 2005, in the laboratory of Dr. Rakesh Jain of Harvard Medical School. During his graduate work, he pioneered the application of new imaging and analysis technologies to studying drug transport within tumors, and on developing methods to improve drug delivery. He also holds a Bachelors of Science in Chemical Engineering with a Biotechnology minor from the University at Buffalo. He moved to Toronto to continue postdoctoral work at the Ontario Cancer Institute, applying multi-modality imaging and quantitative image analysis methods to study preclinical cancer models. He has a successful track record of high-impact publications with a number of clinical and basic science collaborators, and has also collaborated with pharmaceutical companies on imaging-based preclinical testing of new compounds. He is currently Image Analysis Core Manager of the STTARR Innovation Centre, and manages a team of analysts to develop new algorithms for machine-learning powered image segmentation and quantification across a number of disease sites. His research interests lie in studying the tumor microenvironment, drug and oxygen delivery, and the development of tools for “tissue cytometry” – deriving complex biological and spatial relationships from tissue sections via computational image analysis methods.

Large-Scale Analytics and Machine Learning for Biomedical Data Types

Wednesday, June 21st, 2017

Wednesday June 28, 2017 at 5:00 p.m. Dr. Shiva Amiri, CEO of BioSymetrics Inc, will be presenting “Large-Scale Analytics and Machine Learning for Biomedical Data Types”.

Day & Time: Wednesday June 28, 2017
5:00 p.m. – 6:00 p.m.

Speaker: Dr. Shiva Amiri
CEO of BioSymetrics Inc
Toronto, Ontario, Canada

Location: Room ENG288
Department of Computer Science
Ryerson University
245 Church St, Toronto, M5B 1Z4

Contact: Alireza Sadeghian, Alex Dela Cruz

Organizers: Signals & Computational Intelligence Chapter, WIE

Abstract: The scale of data being generated in medicine and research can easily overwhelm typical analytic capabilities. This is particularly true with MRI/fMRI scanning, genomics data, streaming/wearables data in addition to other clinical data types, especially if in combination.

Challenges include 1) large file sizes often in heterogeneous formats 2) currently no standard Protocol exists for extraction of standardized characteristics, and 3) traditional methods for group-wise comparison can often result in spurious findings.

The talk will address these challenges by discussing customized processing pipelines built for multiple data types in biomedicine, which enable effective machine learning and other types of analytics on these datasets. This approach leverages the rapid model building capabilities of our real-time machine learning software to iterate through normalization parameters for each data type and disease class. In addition, this platform allows easy integration between the various medical data types (genome sequence, phenotypic, and metabolic data) allowing generation of more comprehensive disease classification models.

The ability to standardize and pre-process multiple types of biomedical data for machine learning, no matter the source and type, and effectively combine it with other data types is a powerful capability and holds promise for the future of diagnostics and precision medicine.

Biography: Shiva Amiri is the CEO of BioSymetrics Inc. where they are developing a unique real-time machine learning technology for the analysis of massive data in biomedicine. BioSymetrics specializes in providing optimized pipelines for complex data types and effective methods in the analytics of integrated data. Prior to BioSymetrics she was the Chief Product Officer at Real Time Data Solutions Inc., she has led the Informatics and Analytics team at the Ontario Brain Institute, where they developed Brain-CODE, a large-scale neuroinformatics platform across the province of Ontario. She was previously the head of the British High Commission’s Science and Innovation team in Canada. Shiva completed her Ph.D. in Computational Biochemistry at the University of Oxford and her undergraduate degree in Computer Science and Human Biology at the University of Toronto. Shiva is involved with several organisations including Let’s Talk Science and Shabeh Jomeh International.

Engineering the Internet of Things – Digital Twin Seminar

Thursday, April 13th, 2017

Friday April 28, 2017 at 9:00 a.m. IEEE Toronto and SimuTech Group will be hosting the seminar “Engineering the Internet of Things – Digital Twin”.

Day & Time: Friday April 28, 2017
9:00 a.m. – 4:00 p.m.

Location: Ryerson University
George Vari Centre for Computing and Engineering
Room: ENG 288
245 Church Street
Toronto, Ontario M5B 2K3

Cost: Free including lunch

Register: http://go.simutechgroup.com/ieee-iot-digital-twin-toronto

Contact: SimuTech Group – Mohsen Tayefeh
IEEE Toronto – Dr. Maryam Davoudpour

Organizers: IEEE Toronto (WIE, Signals & Computational Intelligence, Measurement/Instrumentation-Robotics, Magnetics chapters), Computer Science Department of Ryerson University, SimuTech Group (ANSYS Elite Channel partner)

Abstract: High-tech–industry product development teams routinely use coupled multiphysics software to analyze the trade-offs among speed, bandwidth, signal integrity, power integrity, thermal performance and EMI/EMC.

The Internet of Things is a network of smart products, or “things”, that use embedded sensors, software, and electronics to communicate with each other over a network. The communication data can be analyzed by cloud based software to derive actionable information, leading to predictive and prescriptive outcomes.

In this seminar, the following topics will be discussed:

– Engineering the Internet of Things
– 5 Engineering Challenges for Smart Product Development
– Case Study: Search and Rescue Drone-Satellite System
– Signal Integrity/EMI/EMC, Human body, Federal Regulations
– User experience – Wearable devices (Multiphysics Simulation)
– Digital Twin – GE and ANSYS collaboration
– Case Study: prescriptive maintenance case study
– Lunch
– RF Antenna placement
– Step by step workshop – Antenna analysis
– PCB design – Power Integrity
– Thermal management (CFD)
– Networking, Door prize/draw (Drone)

Developing Wearable Technologies for improved management of sleep-related breathing disorders

Wednesday, November 23rd, 2016

Tuesday November 29th, 2016 at 2:30 p.m. Dr. Azadeh Yadollahi, Scientist at SleepdB Laboratory and Assistant Professor at University of Toronto, will be presenting “Developing Wearable Technologies for improved management of sleep-related breathing disorders”.

Speaker: Dr. Azadeh Yadollahi
Scientist, SleepdB Laboratory, Toronto Rehabilitation Institute
Assistant Professor, Biomaterial & Biomedical Engineering, University of Toronto
Adjunct Faculty, Department of Biomedical Engineering, University of Manitoba

Day & Time: Tuesday, November 29th, 2016
2:30 p.m. – 3:30 p.m.

Location: Room ENG-460
245 Church Street, Toronto, ON
Ryerson University

Organizer: IEEE Signal Processing Chapter Toronto Section

Contact: Mehrnaz Shokrollahi

Abstract: Over four million Canadians live with a chronic respiratory disease such as asthma, chronic obstructive pulmonary disease (COPD) or obstructive sleep apnea (OSA)—all of which are associated with high morbidity. In Canada, 6.5% of total health care costs are related to these disorders, amounting to $5.7B in direct and $6.72B in indirect costs per year. Moreover, the overlap between asthma, COPD, and OSA is common, is clinically important, worsens quality of life, and is associated with greater morbidity and mortality more than the sum of the contributing disorders. A feature common to chronic respiratory diseases is that their symptoms, eg. shortness of breath, worsen during sleep. Most emergency visits and deaths related to asthma and COPD occur during the night. However, our understanding of the mechanisms of respiratory disorders exacerbation at night is limited; which consequently challenges our ability to manage these disorders. One of the main barriers to determine the underlying pathophysiology of sleep-related respiratory disorders is that the available technologies to perform studies are expensive, invasive, and confound normal breathing and sleep patterns. Therefore, the results may not be applicable to a wide range of people or over a long period of time to evaluate treatments and interventions. Therefore, the mechanistic link between sleep and respiratory disease, particularly the role of night-time fluid redistribution, is not well understood. To address this gap, my team is developing novel technologies to monitor respiratory related physiological signals during sleep, as well and technologies to non-invasively assess tissue composition, and its role on the pathophysiology of sleep related breathing disorders.

Biography: Dr. Azadeh Yadollahi is a Scientist at the Toronto Rehabilitation Institute – University Health Network, where she leads the SleepdB laboratory. She is also an Assistant Professor in the Institute of Biomaterial and Biomedical Engineering, University of Toronto and Adjunct Faculty Member in the Graduate Department of Biomedical Engineering at the University of Manitoba. Her research aims to determine the pathophysiology of sleep-related breathing disorders and to develop novel technologies for improved management of these disorders. She is particularly interested in developing innovative technologies for monitoring of physiological signals at home and implementing personalized treatments for older populations with chronic sleep-related respiratory diseases. To date, Dr. Yadollahi has authored and co-authored more than 30 peer-reviewed publications, had more than 60 presentations at national and international conferences, and been invited 26 times to give presentations on her research at prominent national and international academic institutions. Her research is supported by grants from the Canada Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada (NSERC), Canadian Respiratory Research Network, and Ontario Centres of Excellence, among others. In the past 10 years, Dr. Yadollahi has been instrumental in developing new wearable technologies for improved diagnosis and treatment of breathing disorders during sleep. At Toronto Rehab, Dr. Yadollahi is leading SleepdB, a Sound-proof laboratory to examine sleep-disordered Breathing. SleepdB is the first laboratory in the world dedicated to understanding the mechanisms of airway narrowing during sleep and to developing acoustic technologies to improve sleep-related respiratory disorders. This laboratory will also serve as a hub for knowledge translation and exchange between researchers and clinicians to advance clinically relevant research and implement cutting-edge assessments and treatments for breathing disorders.

Ground Truth Bias in External Cluster Validity Indices

Tuesday, June 21st, 2016

June 28, 2016 at 2:00 p.m. IEEE CIS Distinguished Lecturer James C. Bezdek will be presenting “Ground Truth Bias in External Cluster Validity Indices”.

Speaker: James C. Bezdek
IEEE CIS Distinguished Lecturer

Day & Time: Tuesday, June 28, 2016
2:00 p.m. – 4:00 p.m.

Location: Room ENG 106, George Vary Engineering & Computing Centre
245 Church St., Toronto, ON, M5B 2K3
(Intersection of Church and Gould)

Map: http://www.ryerson.ca/maps/

Contact: Dr. Maryam Davoudpour, Dr. Glaucio Carvalho, Dr. Alireza Sadeghian

Organizers: Signals & Computational Intelligence Chapter, Magnetics Chapter, Instrumentation & Measurement/Robotics & Automation Chapter

Abstract: This talk begins with a short review of clustering that emphasizes external cluster validity indices (CVIs). A method for generalizing external pairbased CVIS (e.g., the crisp Rand and Jacard indices) to evaluate soft partitions is described and illustrated. Three types of validation experiments conducted with synthetic and real world labeled data are discussed: “best c” (internal validation with labeled data), and “best I/E” (agreement between an internal and external CVI pair).

As is always the case in cluster validity, conclusions based on empirical evidence are at the mercy of the data, so the reported results might be invalid for different data sets and/or clustering models and algorithms. But much more importantly, we discovered during these tests that some external cluster validity indices are also at the mercy of the distribution of the ground truth itself. We believe that our study of this surprising fact is the first systematic analysis of a largely unknown but very important problem ~ bias due to the distribution of the ground truth partition.

Specifically, in addition to the well known bias in many external CVIs caused by monotonic dependency on c, the number of clusters in candidate partitions, there are two additional kinds of bias that can be caused by an unusual distribution of the clusters in the ground truth partition provided with labeled data. The most important ground truth bias is caused by imbalance (unequally sized labeled subsets). We demonstrate these effects with randomized experiments on 25 pair-based external CVIs. Then we provide a theoretical analysis of bias due to ground truth for several CVis by relating Rand’s index to the Havrda-Charvat quadratic entropy.

Biography: Jim received the PhD in Applied Mathematics from Cornell University in 1973. Jim is past president of NAFIPS (North American Fuzzy Information Processing Society), IFSA (International Fuzzy Systems Association) and the IEEE CIS (Computational Intelligence Society): founding editor the Int’l. Jo. Approximate Reasoning and the IEEE Transactions on Fuzzy Systems: Life fellow of the IEEE and IFSA; and a recipient of the IEEE 3rd Millennium, IEEE CIS Fuzzy Systems Pioneer, and IEEE technical field award Rosenblatt medals. Jim’s interests: woodworking, optimization, motorcycles, pattern recognition, cigars, clustering in very large data, fishing, co-clustering, blues music, wireless sensor networks, poker and visual clustering. And of course, clustering in big data. Jim retired in 2007, and will be coming to a university near you soon.

Semi-automated Genome Annotation and an Expanded Epigenetic Alphabet

Friday, January 22nd, 2016

Thursday February 11th, 2016 at 1:00 p.m. Michael Hoffman, Principal Investigator at Princess Margaret Cancer Centre and Assistant Professor in the Departments of Medical Biophysics, University of Toronto, will be presenting “Semi-automated genome annotation and an expanded epigenetic alphabet”.

Speaker: Michael Hoffman
Principal Investigator at Princess Margaret Cancer Centre
Assistant Professor in the Departments of Medical Biophysics, University of Toronto

Day & Time: Thursday, February 11, 2016
1:00 p.m. – 2:00 p.m.

Location: Room LG04, George Vari Engineering and Computing Centre
Ryerson University, Toronto, M5B 1Z4
Please check before the seminar

Contact: llivi@scs.ryerson.ca

Abstract: First, we will discuss Segway, an integrative method to identify patterns from multiple functional genomics experiments, discovering joint patterns across different assay types. We apply Segway to ENCODE ChIP-seq andDNase-seq data and identify patterns associated with transcription start sites, gene ends, enhancers, CTCF elements, and repressed regions. Segway yields a model which elucidates the relationship between assay observations and functional elements in the genome.

Second, we will discuss a new method to discover transcription factor motifs and identify transcription factor binding sites in DNA with covalent modifications such as methylation. Just as transcription factors distinguish one standard nucleobase from another, they also distinguish unmodified and modified bases. To represent the modified bases in a sequence, we replace cytosine (C) with symbols for 5-methylcytosine (5mC), 5-hydroxylmethylcytosine (5hmC), 5-formylcytosine (5fC). Similarly, we adapted the well-established position weight matrix model of transcription factor binding affinity to an expanded alphabet. We created an expanded-alphabet genome sequence using genome-wide maps of 5mC, 5hmC, and 5fC in mouse embryonic stem cells. Using this sequence and expanded-alphabet position weight matrixes, we reproduced various known methylation binding preferences, including the preference of ZFP57 and C/EBPβ for methylated motifs and the preference of c-Myc for unmethylated motifs. Using these known binding preferences to tune model parameters enables discovery of novel modified motifs.

Biography: Michael Hoffman is a principal investigator at the Princess Margaret Cancer Centre and Assistant Professor in the Departments of Medical Biophysics and Computer Science, University of Toronto. He researches the application of machine learning techniques to epigenomic data. He previously led the National Institutes of Health ENCODE Project’s large-scale integration task group while at the University of Washington. He has a PhD from the University of Cambridge, where he conducted computational genomics studies at the European Bioinformatics Institute. He also has a B.S. in Biochemistry and a B.A. in the Plan II Honors Program at The University of Texas at Austin. He was named a Genome Technology Young Investigator and has received several awards for his academic work, including a NIH K99/R00 Pathway to Independence Award.

Every Picture Tells a Story: Visual Cluster Assessment in Square and Rectangular Relational Data

Sunday, November 8th, 2015

Monday December 7, 2015 at 4:00 p.m. Professor Emeritus James Bezdek will be presenting “Every Picture Tells a Story: Visual Cluster Assessment in Square and Rectangular Relational Data”.

Speaker: Emeritus James Bezdek
Past President of NAFIPS, IFSA and the IEEE CIS

Day & Time: Monday, December 7, 2015
4:00 p.m. – 6:00 p.m.

Location: Room 1180
Bahen Center for Information Technology
40 St. George Street, Toronto

Organizer: IEEE Toronto Signals & Computational Intelligence Chapter
Distinguished Lecturer Program

Contact: Lorenzo Livi, Email:llivi@scs.ryerson.ca

Abstract: The VAT/iVAT, algorithms are the parents of a large family of visual assessment models.

Part 1. Definitions of the three canonical problems of cluster analysis: tendency assessment, clustering, and cluster validity. History of Visual Clustering. Applications: role-based compliance assessment, eldercare time series data, and anomaly detection in wireless sensor networks.

Part 2. Extension to siVAT, scalable iVAT for big data. This is the basis of clusiVAT and clusiVAT+ for clustering in big data (Topic 4 below). Application: image segmentation. Extension to coiVAT for assessment of co-clustering tendency in the four clustering problems associated with rectangular relational data. Application: response of 18 Fetal Bovine Serum Treatments to the treatment of fibroblasts in gene expression data.

Biography: Jim received the PhD in Applied Mathematics from Cornell University in 1973. Jim is past president of NAFIPS (North American Fuzzy Information Processing Society), IFSA (International Fuzzy Systems Association) and the IEEE CIS (Computational Intelligence Society): founding editor the Int’l. Jo. Approximate Reasoning and the IEEE Transactions on Fuzzy Systems: Life fellow of the IEEE and IFSA; and a recipient of the IEEE 3rd Millennium, IEEE CIS Fuzzy Systems Pioneer, and IEEE technical field award Rosenblatt medals. Jim’s interests: woodworking, optimization, motorcycles, pattern recognition, cigars, clustering in very large data, fishing, co-clustering, blues music, wireless sensor networks, poker and visual clustering. And of course, clustering in big data. Jim retired in 2007, and will be coming to a university near you soon.

Learning in Non-stationary Environments

Sunday, August 30th, 2015

October 6, 2015 at 11:00 a.m. Cesare Alippi, IEEE Fellow & Professor of Information Processing Systems with the Politecnico di Milano, will be presenting a distinguished lecture, “Learning in Non-stationary Environments” at Ryerson University.

Speaker: Cesare Alippi
IEEE Fellow
Professor of Information Processing Systems with the Politecnico di Milano

Day & Time: Tuesday, October 6, 2015
11:00 a.m. – 12:00 p.m.

Location: George Vari Centre for Computing and Engineering
Ryerson University
Room: ENG287
245 Church Street, Toronto, Ontario M5B 2K3
Click here to see the Map – Look for ENG

Organizer: IEEE Signals & Computational Intelligence Toronto Chapter

Contact: E-mail: Lorenzo Livi

Abstract: Most of machine learning applications assume the stationarity hypothesis for the process generating the data. This amenable assumption is so widely –and implicitly- accepted that sometimes we even forget that it does not generally hold in the practice due to concept drift (i.e., a structural change in the process generating the acquired datastreams). The ability to detect concept drift and react accordingly is hence a major achievement for intelligent learning machines and constitutes one of the hottest research topics for embedded systems. This ability allows the machine for actively tuning the application to maintain high performance, changing online the operational strategy, detecting and isolating possible occurring faults to name a few relevant tasks. The talk will focus on “Learning in a non-stationary environments”, by introducing both passive and active approaches. The active approach will be deepened by presenting triggering mechanisms based on Change point methods and Change detection tests. Finally, the just-in-time detect&react mechanism is introduced where, following a detected change, the system immediately reacts with a strategy depending on the available information.

Biography: Cesare Alippi received the degree in electronic engineering cum laude in 1990 and the PhD in 1995 from Politecnico di Milano, Italy. Currently, he is a Full Professor of information processing systems with the Politecnico di Milano. He has been a visiting researcher at UCL (UK), MIT (USA), ESPCI (F), CASIA (RC), A*STAR (SIN).
Alippi is an IEEE Fellow, Distinguished lecturer of the IEEE CIS, Member of the Board of Governors of INNS, Vice-President education of IEEE CIS, Associate editor (AE) of the IEEE Computational Intelligence Magazine, past AE of the IEEE-Trans. Instrumentation and Measurements, IEEE-Trans. Neural Networks, and member and chair of other IEEE committees.
In 2004 he received the IEEE Instrumentation and Measurement Society Young Engineer Award; in 2013 he received the IBM Faculty Award. He was awarded the 2016 IEEE TNNLS outstanding paper award.
Among the others, Alippi was General chair of the International Joint Conference on Neural Networks (IJCNN) in 2012, Program chair in 2014, Co-Chair in 2011. He was General chair of the IEEE Symposium Series on Computational Intelligence 2014.
Current research activity addresses adaptation and learning in non-stationary environments and Intelligence for embedded systems.

Alippi holds 5 patents, has published in 2014 a monograph with Springer on “Intelligence for embedded systems” and (co)-authored more than 200 papers in international journals and conference proceedings.
Home Page: http://home.dei.polimi.it/alippi/