About the talk
Dr. Ariane Marelli is Professor of Medicine at McGill University. She is the Cardiovascular Program Leader at the Research Institute of the McGill University Health Centre and the Associate Director of Academic Affairs for Cardiology at the McGill University Health Centre. She completed her training in adult cardiology and a fellowship in pediatric cardiology at McGill University. She obtained advanced fellowship training in adult congenital heart disease at the University of California in Los Angeles where she was the first person to graduate from the first adult congenital heart disease program in the US. She obtained an MPH at the Harvard School of Public Health. She came to the McGill University Health Centre in l997 where she founded and directs the McGill Adult Unit for Congenital Heart Disease (MAUDE Unit) that now follows over 3 000 patients. She is the only Canadian to serve on the Epidemiology and Prevention Council, Statistics Committee of the American Heart Association where she is a member of the Scientific Program Committee. She was appointed to the leadership council of the American College of Cardiology, Pediatric Cardiology and Adult Congenital Section where she co-chairs the Quality Working Group. She is President of the Canadian Adult Congenital Heart Network.View the profile
His current research efforts are to develop Green Chemistry for organic synthesis based upon innovative and fundamentally new organic reactions that will defy conventional reactivities and possess high “atom-efficiency”. Well-known researches developed by him include the development of a wide range of Grignard-type reactions in water, transition-metal catalysis in air and water, alkyne-aldehyde-amine coupling (which he termed A3-coupling), and cross-dehydrogenative-coupling (which he formulated and termed CDC) reactions among others, biomass conversion to chemicals, light-assisted reactions, methane/natural gas conversion to hydrogen and aromatics, solar-enabled nitrogen to ammonia conversions, conversions of lignin phenols into anilines, direct conversion of methanol into ethanol, and umpolung to carbonyls into organometallic reagent surrogates. His research has been cited widely in the literature (>47,000 times, h-index=105) and was featured as one of the top 20 Canadian Chemistry Discoveries in the past century by the Canadian Chemical News in 2007 as well as on the lists of the World Most Highly Cited Scientists by Thomson Reuters (2014, 2015, 2016, 2017). He has published >500 peer-reviewed articles and has given >470 plenary/keynote/invited lectures and received a number of prestigious awards/honors worldwide.View the profile
Right after finishing my Ph.D study, I joined the Verglas Project as a statistician. Mainly my work is to manage data, analyze data, interpret results, and involve in writing manuscripts. Be doing this, I work with my supervisors and colleagues in the project together to investigate the influence of prenatal stress on the physical growth, functioning, cognitive development, and behavior of the children exposed to such stress. My first post-secondary education was completed at Beijing Medical University in China, where I received a bachelor degree of medicine with a focus on preventive medicine. Motivated by the desire to improve myself and my career, I came to McGill University to start my Master study in the Department of Epidemiology and Biostatistics, and then a Ph.D study in the same department. During my Ph.D study, I had been focusing on the methodological issues of testing and estimating interactions.View the profile
Jay Brophy began academic life as a chemical engineer, completed medical school at McMaster University, did an internal medicine and cardiology residency at the University of Montreal, and completed a PhD in Epidemiology & Biostatistics at McGill University. He is a professor in the Departments of Medicine and Epidemiology & Biostatistics & Occupational Health and is a practicing cardiologist at the McGill University Health Centre. His research interests are eclectic involving cardiovascular medicine, pharmacoepidemiology and drug safety, medical decision making, as well as health technology assessments including economic analyses. He is a funded scholar from les Fonds de Recherche Quebec (Santé) and has published over 300 peer reviewed publications.View the profile
David Buckeridge is a Professor of Epidemiology and Biostatistics at McGill University in Montreal where he holds a Canadian Institutes of Health Research Chair in e-Health Interventions. He is also a Medical Consultant to the Montreal Public Health Department and the Quebec Public Health Institute. Dr Buckeridge has consulted on surveillance to organizations such as the Public Health Agency of Canada, the US Institute of Medicine, the US and Chinese Centers for Disease Control, the European Centers for Disease Control, and the World Health Organization. He holds a M.D. from Queen’s University, a M.Sc. in Epidemiology from the University of Toronto, a Ph.D. in Biomedical informatics from Stanford University and is a Fellow of the Royal College of Physicians and Surgeons of Canada with specialty training in Public Health and Preventive Medicine.View the profile
I'm currently an assistant professor at HEC Montreal (Business school of University of Montreal) and Montreal Institute for Learning Algorithms, which is the institute focusing on deep learning and reinforcement learning lead by one of the deep learning pioneers Yoshua Bengio. My research focuses on deep learning and reinforcement learning with applications on graph representation learning, recommender systems, and natural language understanding.View the profile
From 2015 to 2018, I was a postdoctoral associate from Prof. Manolis Kellis research group at Computer Science and Artificial Intelligence Laboratory (CSAIL) at Massachusetts Institute of Technology. I obtained a PhD degree in Computer Science and Computational Biology at University of Toronto in 2014. My PhD advisor was Prof. Zhaolei Zhang. I received Bachelor of Science Honors from the University of Saskatchewan in Computer Science Bioinformatics and Statistics in 2010. My honor thesis advisor was Prof. Anthony Kusalik.View the profile
Hi, my name is Han Newton and I will be presenting our project entitled machine. Learning automation is SHINee, designed empirical manual for congenital, heart disease. Identification. Enlarged veins databases. Congenital heart disease is an umbrella term for the newest heart defect a quite a bit worse. Patients with congenital heart disease are surviving longer into adulthood thanks to the advances of man is long as the Bible along with the increased. Number of patients allows for better research in this patient population, large twins databases are an excellent
source of research data as it provides very large samples, with little to no referral bias, the first step and challenge in using them. In congenital heart disease, research is the ability to identify shoot, General hide his patients, as opposed to a patient who has a diagnosis of congenital heart disease in their file, but you don't actually have the disease. In fact, General heydrich is a very complex diagnosis to me, as it has 25 possible, lesion with varying degrees of severity. So, at the correct, diagnosis, was quite expensive. And time-consuming manual chart
review to try to solve this problem. We sought to develop machine learning algorithm capable of extracting patient. Need a large administrative and Healthcare databases in Quebec, Canada and compare it to clinician developed. 19187 patients with at least one diagnosis of congenital heart disease made by an on cardiovascular Specialists between the nearest 1983. Mm, we reviewed these 3784 patient with previously, identified 290 audit as true congenital heart disease patients. We committed to study population to to
prediction model. A conventional logistic regression. On the one hand and machine learning models to evaluate their performance on identifying truth. Gentle heart disease. Patients machine learning algorithm, sorry. The machine Learning Arm was further, subdivided into a train set and a text in between two machine learning model named re-entry and support Vector machine with the training date of that. Using a binary outcome. Yes and no heart disease features. We used to construct a model.
The results with that decision tree followed by support Vector machine African conventional logistic regression incorrect. Identification of congenital heart disease patient with both accuracy sensitivity and specificity and externally validated the decision tree with a much larger sample of 68000 patients and they showed maintenance of the excellent accuracy of a complex disease. From a large administrative database and automated in a sitting position time and energy. Thank you.
Buy this talk
Buy this video
Our other topics
With ConferenceCast.tv, you get access to our library of the world's best conference talks.