U of T Technology & Startup Explorer

Machine learning architecture offers a novel, high-fidelity means of assessing lung physiology during the performance of (monofrequency) oscillometry tests.

Pulmonary Embolism (PE) diagnosis using machine learning (ML) techniques on CT pulmonary angiography (CTPA) is transforming the field of medical imaging. PE is a life-threatening condition where blood clots block the pulmonary arteries, requiring timely and accurate detection to improve patient outcomes. Traditional training and development of ML models for PE diagnosis involve extensive annotation, which is resource-intensive. To make these applications practical, the team has developed optimized training strategies, model architectures, and deployment pipelines for efficient PE detection.

Machine-learning based methods to improve traffic flow through adaptive signal control.

Serum-free cell culture media design is important because it offers the opportunity to construct chemically defined media that possess greater batch-to-batch reproducibility and clinically compatible reagents; benefits not necessarily associated with serum-containing media.  As such, it is an ideal method for producing media for the growing cell therapy market segment where high quality and reproducibility standards influence clinical translation.  However, the use of defined chemical factors often creates a complex, large-scale optimization problem where the number of possible combinations of factors across different dose concentrations creates a search space so large that only a millionth or less of it can be sampled.  Consequently, this renders traditional media optimization techniques ineffective and necessitates the development of new techniques to handle this problem.

AI-powered cyber threat detection honeypot technologies for ICS, CPS, and IoT that actively engage attackers and draw out more information about their tactics, providing organizations with the intelligence they need to better defend their systems.

Toronto Annotation Suite (TORAS) is a web-based AI-powered annotation platform.

Hippocamera is a memory augmentation technology that aides in the recall and storage of memories. The technology features novel methods to tag and replay recorded moments to promote rich reliving and storage of experiences.

JALI provides products and services for the complete automation of high-end lip sync and facial animation for ultimate animator directorial control.

ODAIA is business intelligence software that enables organizations to understand, predict, and optimize their customer's journeys. The technology leverages AI to facilitate the exploration, analysis, summarization, and optimization of a customer's journey, from awareness to purchase.

This opportunity is a portfolio of hardware and software techniques that seamlessly accelerate AI/ML applications, while saving on-chip and off-chip memory, bandwidth, and processing power. These compute and compression methods have been shown to deliver a 3x boost in processing speed with 2x-4x increase in storage capabilities (up to 10x for natural language processing).

High-performance software solutions and advanced algorithms for cryo-EM data processing.

AutoScribe uses hands-free speech recognition and Artificial Intelligence to help reduce time consuming manual clinical documentation, freeing clinicians to place their focus where it should be – on their patients.

A method to extract, assess, and visualize information from electroencephalography (EEG) readings.

Electronic Systems for Monitoring, Diagnosis, and Control of Physiological Disorders

The two inventions captured in this technology allow for the representation, learning, and interactive editing of color palettes through a novel interface for creating color themes.