
Microsoft debuts an advanced diagnostic AI model, while major studies highlight both technical progress and patient trust concerns in medical AI.
Key Details
- 1Microsoft's MAI-DxO AI outperformed physicians on 85% of 304 NEJM real-world cases, reportedly quadrupling expert diagnostic rates.
- 2MIT and Mass General Brigham announced a new Analog Devices-funded program for joint medical AI research, aiming for six projects yearly.
- 3MIT researchers found LLMs may misdirect patients using colloquial symptom descriptions, with particular risk for female patients.
- 4A survey for ModMed found 55% of patients are uncomfortable with AI in diagnosis/treatment, yet 57% support it if it improves doctor-patient face time.
- 5New studies confirm patients perceive less empathy from AI-generated responses than from humans, impacting trust.
- 6Recent imaging AI highlights include tools for lung tumor segmentation, dementia diagnosis, and fatty liver detection via radiographs.
Why It Matters

Source
AI in Healthcare
Related News

Toronto Study: LLMs Must Cite Sources for Radiology Decision Support
University of Toronto researchers found that large language models (LLMs) such as DeepSeek V3 and GPT-4o offer promising support for radiology decision-making in pancreatic cancer when their recommendations cite guideline sources.

AI Model Using Mammograms Enhances Five-Year Breast Cancer Risk Assessment
A new image-only AI model more accurately predicts five-year breast cancer risk than breast density alone, according to multinational research presented at RSNA 2025.

AI Model Uses CT Scans to Reveal Biomarker for Chronic Stress
Researchers developed an AI model to measure chronic stress using adrenal gland volume on routine CT scans.