As Dr Paul Chang (one of Radiology's "20 Most Influential People") recently remarked, "Healthcare industries don't have the proper IT infrastructure to feed and consume AI and deep learning data." There is incredible promise in applying AI to healthcare, particularly imaging. We constantly see reports of AI algorithms performing a task well enough to augment radiologists' abilities. Yet the details of how to bring these amazing breakthroughs to everyday clinical practice are still being worked out. A robust, low-latency IT infrastructure is key to integrating these advancements in AI into existing workflows in such a way that they help both physicians and patients.

Key points covered include:

• How the proper infrastructure can enable the translation of AI from the lab to clinical care
• How latency can negate the benefits of AI in a patient-care setting
• How to avoid "GPU starvation"