Enhance physician workflow with state-of-the-art AI models, for research, seamlessly integrated with your existing PACS.
?Enhance your diagnostic capabilities with our comprehensive AI-powered platform.
Use AI without infrastructure overhaul.
Replace your PACS browser with AI-powered insights.
Ideal for testing clinical AI applications in the real world for research use or run regulated AI models.
Run open-source (for research use) or regulated AI models directly on local PACS.
Self-deployment designed for privacy, compliance, and ease
Our containerized architecture ensures rapid deployment, consistent performance, and seamless scalability.
PACS-AI uses docker containers for rapid deployment and consistent environments.
Plug-and-play AI models for different imaging modalities.
End-to-end encryption data handling.
Publications
Explore studies and preprints behind models developed by or available through PACS-AI.
Available at no cost to clinicians, researchers, and institutions evaluating AI in medical imaging.
Ideal for academic and research institutions exploring AI in medical imaging.
Research license
Access to research-focused AI models
Data collection tools
Research collaboration support
Academic publication rights
Technical documentation access
* Non-FDA/Health Canada approved models. For research purposes only.
Need more details? Get in touch with our team!
PACS-AI seamlessly integrates with your existing PACS infrastructure, enabling AI-powered image analysis without requiring a major system overhaul. It supports both local AI model testing for researchers and full PACS browser replacement for clinicians, enhancing diagnostic workflows.
PACS-AI is currently undergoing the regulatory assessment to see if Health Canada approval is needed. While awaiting regulatory clearance, it is available for research and non-diagnostic use after local review board approval, under an approved research protocol, with a full clinical version planned upon approval.
Yes! PACS-AI offers a self-hosted version, allowing hospitals and research institutions to deploy it within their infrastructure for full control over data security and compliance. We collect usage analytics (number of users, AI models and number of AI queries for self-hosted versions but do not store any patient data or any AI prediction result).
PACS-AI follows a hybrid mode:
Yes, anonymous aggregated usage statistics are collected to help improve performance, troubleshoot issues, and enhance user experience. However, our company will not store or share patient data, ensuring full compliance with HIPAA, PIPEDA, and Law 25 in Quebec.
Absolutely! PACS-AI provides a flexible AI inference pipeline, allowing researchers to deploy and validate their own models directly within their local PACS environment, without needing external cloud processing.
PACS-AI includes a built-in PACS viewer with AI-assisted image interpretation, making it a powerful alternative to traditional PACS browsers.
PACS-AI does not sell or share patient data. All data processing happens locally or within a secure on-premises environment. If cloud-based AI inference is used, encryption protocols and compliance measures ensure full data protection.
PACS-AI is developed by a CIFAR (Canadian Institute for Advanced Research) team of clinicians, AI researchers, and engineers committed to advancing medical imaging through AI-driven solutions.
© 2026 HeartWise AI Lab, Montreal Heart Institute. All rights reserved.
