The U of A launched Canada’s first computing science department in April 1964. Over the decades, the university has overcome one artificial intelligence challenge after another to become a world leader in AI research.
Today, U of A scientists across disciplines are advancing what AI can do and how well it can do it. They’re working on fundamental research to advance AI as well as ways to apply AI to health care, energy systems, law, agriculture, smart construction, autonomous vehicles and more. And they are striving to make sure AI works for us, by examining questions of privacy, security, ethics and bias to help us navigate its ever-advancing capabilities.
Smarter Energy Systems
Electrical and computer engineer Hao Liang is leveraging information and communication technology to develop “intelligent” energy systems that are more efficient, reliable, sustainable and secure, including energy systems that use various energy sources.
Fairer Algorithms
Computing scientist Nidhi Hegde, ’95 BSc(Spec), focuses on identifying where and how bias occurs in machine learning and how to build fairness and privacy protection into algorithms to make them more trustworthy, effective and fair for all no matter what race, ethnicity, gender, age or other factors.
Chatbot for Seniors
Computing scientist Osmar Zaiane leads a project involving colleagues in psychiatry that’s exploring how to create an empathetic and emotionally intelligent chatbot companion for seniors. The generative AI tool is also designed to detect signs of depression and dementia and pass that information on to caregivers and health-care providers.
Playlists for Patients
Music professor Michael Frishkopf and his interdisciplinary research team are using machine learning to select soundscapes to reduce stress in intensive care patients. An algorithm assesses a patient’s psychological state by monitoring biosignals and responds with personalized soothing sounds.
Illness Early Warning
A research team led by Sunil Kalmady Vasu, a machine learning specialist in the Faculty of Medicine & Dentistry, has found a way to assess the chances that first-degree relatives of people with schizophrenia will develop the disease, given that they have a risk of up to 19 per cent compared with one per cent in the general population. The tool is not meant to replace diagnosis by a psychiatrist but could lead to earlier diagnosis.
Better Wildfire Responses
Using 15 years of wildfire fighting data, a team led by business professor Ilbin Lee did an experiment with a machine-learning simulation to determine how the resources used in initial attack operations affected success. The results offered insight into the best ways to allocate limited resources to fight wildfires under different conditions.
Gender Bias Repair
Statistics professor Bei Jiang, ’08 MSc, is working on a better way to reduce gender bias in natural language processing models while preserving vital information about the meanings of words. For example, when considering a word like “nurse,” the research team wants the system to remove any gender information associated with that term while retaining information that links it with related words such as doctor, hospital and medicine.
Better Buildings
Researchers in science and engineering are working to optimize building systems through the use of artificial intelligence to improve the planning, design, construction, operation and maintenance of buildings. The goal is to increase the comfort of occupants and reduce energy consumption and costs.
Death Risk Indicator
A team led by Padma Kaul, ’00 PhD, in the Department of Medicine trained a machine-learning algorithm to predict a patient’s risk of death from all causes one month, one year and five years after having an electrocardiogram in hospital. The result was an 85 per cent accuracy rate based on 1.6 million ECGs done on 244,077 patients in northern Alberta between 2007 and 2020. When factors such as age, sex and six standard laboratory blood test results were included, the predictions were even more accurate.
Expert Answers
Augustana computing scientist Mi-Young Kim and colleagues are developing AI that can answer medical and legal questions — and explain its answers at the same time. In collaboration with Alberta Health Services and startup Jurisage, the tool will make hard-to-reach expertise more accessible and save time and money.
Tools for Construction
Aminah Robinson Fayek is an expert in fuzzy logic, an AI technique that represents expert knowledge and subjective reasoning through mathematical models. She is harnessing fuzzy logic, machine learning and simulation to capture expert construction knowledge, which can then be used to improve the accuracy and efficiency of decision-making in construction planning, execution and control.
Addiction Prediction
A team led by Bo Cao, Canada Research Chair in Computational Psychiatry, has created a machine-learning model that can predict with 86 per cent accuracy patients who are at high risk of developing opioid use disorder. Based on nearly 700,000 Alberta patients who received prescriptions for opioids between 2014 and 2018, the project found the top risk factors included frequency of opioid use, high dosage and a history of other substance use disorders.
Responsive Robots
Mechanical engineering professor Ehsan Hashemi is collaborating with colleagues in the Department of Psychology to control networked robots to work safely side by side with humans in dynamic work environments by responding to cues in body language, a branch of experimental psychology.
We at New Trail welcome your comments. Robust debate and criticism are encouraged, provided it is respectful. We reserve the right to reject comments, images or links that attack ethnicity, nationality, religion, gender or sexual orientation; that include offensive language, threats, spam; are fraudulent or defamatory; infringe on copyright or trademarks; and that just generally aren’t very nice. Discussion is monitored and violation of these guidelines will result in comments being disabled.