While scientists rapidly develop our understanding of gastrointestinal (GI) and neurological disorders, these advances can make diagnoses more complex and time intensive. Given the overall burden of these diseases, how can healthcare providers leverage the power of Artificial Intelligence to reduce turnaround times for testing and ultimately improve patient outcomes?
Huron Digital Pathology (Huron) is helping to address this question with a group of researchers at the University of Toronto (U of T) in collaboration with researchers from the University Health Network (UHN), St. Michael’s Hospital, Queen’s University, and the University of Waterloo. The group will scale an existing Atlas of Digital Pathology database, jointly developed by Huron and U of T, to include a range of healthy and diseased GI and neurological tissues. It will then train Artificial Intelligence (AI) and Machine Learning (ML) algorithms on image recognition and annotation and engage pathologists from the participating hospitals to verify results. From there, the project aims to develop a suite of Computational Aided Diagnostic (CAD) tools that can be efficiently deployed in hospitals in Ontario and worldwide.
In September 2022, the group was awarded $1,477,381 (approx. $1.5 million) by the Ministry of Colleges and Universities Ontario Research Fund - Research Excellence (ORF-RE) program.
“The project delivers innovation and addresses current challenges by synergistically integrating pathology domain knowledge with cutting-edge AI research. It builds on, adapts, extends, and scales up AI methodologies for digital pathology developed and successfully demonstrated by the group. It will help build a cost-effective workflow for Ontario hospitals.” says Professor Konstantinos N. Plataniotis, the principal investigator of the study and Director of the Multimedia Lab and Professor at U of T’s Department of Electrical and Computer Engineering.
The project ensures that new tools are suited for clinical environments. For example, screening technology may identify quality issues with slide preparation or imaging. At the same time, a triage tool will help identify initial assessments and gather similar cases to reduce turnaround times for reports, reducing hospital stays. Ultimately, the project aims to create a universally intelligent diagnostic system that can classify the type and severity of a disease based on tissue presentation.
“To uncover the potential of AI in the clinical realm, it is crucial to develop its understanding of the range of ‘normal’ so that it can learn to differentiate normal from the ‘abnormal’ regions on any tissue biopsy. We are undertaking this mammoth task using clinical patients’ GI and Neurological tissue samples. This is like training AI as you would start with a 1st-year pathology resident doctor. Bringing AI into routine clinical practice has implications for making healthcare more efficient and broadly accessible and providing expert care to remote locations. This has never been more imperative than in the current healthcare climate with increasing workloads, burn-out and limited resources.” says Sonal Varma MD, Associate Professor at Queen’s University School of Medicine, Department of Pathology and Molecular Medicine.
Huron Digital Pathology, a provider of digital pathology solutions, will contribute funds and expertise to the project by focusing on how AI can offer novel applications and services for clinicians. As Canada’s only manufacturer of digital scanners for pathology, this project represents a significant opportunity for the country’s participation in global healthcare technology. “Huron is very excited to participate in this project. Huron’s mission is to use its advanced technology to develop scanners and AI software tools to help pathologists identify cancerous diseases and increase efficiency in their diagnosis. This project advances Huron’s mission.” Savvas Chamberlain, Chairman of the Board of Huron Digital Pathology.
Project's Vision
Our goal is to integrate artificial intelligence in developing infrastructure and work-flows for digital diagnostic pathology, facilitating transformational changes in the detection and treatment of disease, improving patient safety, and thus promoting sustainability of the Ontario health system.
Spotlight on Research: Using AI for Digital Pathology
A university-industry-government partnership is developing new artificial intelligence/machine learning tools for the next generation of digital pathology scanners. The project aims to revolutionize pathology infrastructure in Ontario hospitals, facilitating transformational changes in the detection and treatment of disease, improving patient safety, and promoting sustainability of the Ontario health system.
>The project has three main research objectives:
- Creation of an open, flexible, and comprehensive digital pathology database using AI-empowered annotating workflow to scale up and enlarge UofT's currently available database. The database will cover a diverse spectrum of gastrointestinal and neuropathological organs and will be used to learn the visual representation of histological and histopathological components of organs.
- Research, development, validation, and deployment of efficient and customizable convolution neural networks (CNNs) to classify tissue patches extracted from WSI as being histological or histopathological tissue organs. This will allow the development of an AI-based, automated slide-level screening technology, which will alleviate the burden of long turnaround time for test reports and significantly reduce lengthy hospital stays.
- Creation and development of an AI-powered, efficient, and universal disease diagnosing system that can classify types and levels of diseases, identify clinically relevant organ information, and generate relevant patient statistics reports. A preliminary diagnosis will be rendered by ScannerAI to be validated by the attending pathologists.
The project's thorough validation will take place at St. Michael's Hospital in Toronto and Toronto General Hospital/Research Institute (UHN). Commercial and social benefits are expected to accrue from the scanning systems and software products developed. Huron Digital Pathology will provide the means for developing, commercializing, and applying the research outcomes.
Check out our News Release!
</div>