Study Finds Learning Strategies Can Improve AI Model Adaptability in Greater Toronto Area Hospitals

A study published in *JAMA Network Open* identifies specific learning strategies that can improve the effectiveness of artificial intelligence (AI) models used in hospitals. Researchers from York University examined how discrepancies between training data and real-world data could lead to patient harm when AI systems are deployed in medical settings, such as hospitals across the Greater Toronto Area. The study highlights proactive, continual, and transfer learning strategies as critical tools for addressing these data shifts and minimizing potential risks.

The research emphasizes that AI models trained on static datasets may struggle to adapt to evolving real-world conditions, particularly in healthcare environments where patient demographics or clinical practices frequently change. Proactive learning involves anticipating changes in data patterns and preparing AI systems accordingly. Continual learning allows models to update their knowledge over time as new information becomes available, while transfer learning enables them to apply insights gained from one dataset to another with similar characteristics. These approaches aim to enhance the reliability of AI systems and reduce the likelihood of errors stemming from mismatched or outdated training data.

Newsflash | Powered by GeneOnline AI

Source: GO-AI-ne1

Date: June 4, 2025

Author

Mark Chiang