Eight University of Toronto artificial intelligence researchers – four of whom are women – have been named CIFAR AI Chairs, a recognition of pioneering work in areas that could have global societal impact. One of the new chairs is Anna Goldenberg, an associate professor of computer science in U of T's Faculty of Arts & Science and the first-ever chair in biomedical informatics and artificial intelligence at the Hospital for Sick Children. She and her colleagues, including U of T's Dr. Peter Laussen, have developed a computer model that uses signals in physiological data, such as a patient's pulse, to detect an oncoming heart attack – giving doctors and nurses vital minutes to intervene and save an infant's life. The early-warning system has been able to predict 70 per cent of heart attacks at least five minutes – and up to 15 minutes – before a patient's heart stops beating. "In machine learning and health care, the key word is prevention," says Goldenberg, whose team is on track to have the system tested in a silent trial in a clinical environment.

Already, at Massachusetts General Hospital in Boston, "every one of the 50,000 screening mammograms we do every year is processed through our deep learning model, and that information is provided to the radiologist," says Constance Lehman, chief of the hospital's breast imaging division. In deep learning, a subset of a type of artificial intelligence called machine learning, computer models essentially teach themselves to make predictions from large sets of data. The raw power of the technology has improved dramatically in recent years, and it's now used in everything from medical diagnostics to online shopping to autonomous vehicles. But deep learning tools also raise worrying questions because they solve problems in ways that humans can't always follow. If the connection between the data you feed into the model and the output it delivers is inscrutable -- hidden inside a so-called black box -- how can it be trusted?

In clinics around the world, a type of artificial intelligence called deep learning is starting to supplement or replace humans in common tasks such as analyzing medical images. Already, at Massachusetts General Hospital in Boston, "every one of the 50,000 screening mammograms we do every year is processed through our deep learning model, and that information is provided to the radiologist," says Constance Lehman, chief of the hospital's breast imaging division. In deep learning, a subset of a type of artificial intelligence called machine learning, computer models essentially teach themselves to make predictions from large sets of data. The raw power of the technology has improved dramatically in recent years, and it's now used in everything from medical diagnostics to online shopping to autonomous vehicles. But deep learning tools also raise worrying questions because they solve problems in ways that humans can't always follow.

We study prioritized planning for Multi-Agent Path Finding (MAPF). Existing prioritized MAPF algorithms depend on rule-of-thumb heuristics and random assignment to determine a fixed total priority ordering of all agents a priori. We instead explore the space of all possible partial priority orderings as part of a novel systematic and conflict-driven combinatorial search framework. In a variety of empirical comparisons, we demonstrate state-of-the-art solution qualities and success rates, often with similar runtimes to existing algorithms. We also develop new theoretical results that explore the limitations of prioritized planning, in terms of completeness and optimality, for the first time. This paper was published at AAAI 2019.

Dr. Anna Goldenberg, senior scientist in genetics and genome biology at SickKids, poses at the Peter Gilgan Centre for Research and Learning in Toronto. Inside the pediatric intensive care unit at Toronto's Hospital for Sick Children, an infant recovering from open-heart surgery is barely visible through the forest of whizzing and beeping machines that monitor his every vital sign. In the old days, those vital signs – a baby's heart rate, blood pressure, oxygen levels and other signals – would have flashed across a screen and then been lost to posterity. But in 2013, SickKids began collecting and storing the data that emanate from patients in their 42 intensive-care beds. The unit now has more than two trillion data points in its virtual vault, far more than a mere mortal could make sense of.

At a large technology conference in Toronto this fall, Anna Goldenberg, a star in the field of computer science and genetics, described how artificial intelligence is revolutionizing medicine. Algorithms based on the AI principle of machine learning now can outperform dermatologists at recognizing skin cancers in blemish photos. They can beat cardiologists in detecting arrhythmias in EKGs. In Goldenberg's own lab, algorithms can be used to identify hitherto obscure subcategories of adult-onset brain cancer, estimate the survival rates of breast-cancer patients, and reduce unnecessary thyroid surgeries. It was a stunning taste of what's to come.

Letting a snake-like robot glide into your mouth and down your throat may sound a bit alarming. Letting such a robot glide into any of your other orifices may sound more alarming still. But the Flex Robotic System from Medrobotics, in Raynham, Mass., has earned high praise from a head and neck surgeon who has sent it snaking down 19 of his patients' throats as of today. "It really is changing the way I do business," says David Goldenberg, director of otolaryngology surgery at the Penn State Hershey Medical Center. "It is the future of head and neck surgery."