In: Machine Learning and Artificial Intelligence

A close-up photo of a doctor taking notes on a clip board.

In others’ words: Using large language models to accurately analyze doctors’ notes and improve the reliability of real-world AI applications

Johns Hopkins and Columbia University computer scientists teamed up to combat the inaccurate correlations that artificial intelligence and machine learning models learn from text data.

The Circlage dashboard on a MacBook Pro laptop screen.

AI, the new surgical mentor

A collaboration with researchers at the University of Maryland, Baltimore, Circlage is a cloud-based surgical video analysis platform designed to standardize, critique, and train surgeons on designated procedures.

Four overlays of a simulated pelvic X-ray image. The top left has no overlay. The top right has anatomical landmarks labeled. The bottom left shows semantic segmentation annotations for the bones and orthopedic hardware. The bottom right are more specific segmentations for bony corridors that the procedure is targeting.

X-ray vision: Tech could improve efficiency of pelvic fracture surgery

Johns Hopkins researchers harness the power of machine learning to develop a first approach to X-ray-guided surgical phase recognition.

3D render illustration of a group of tablets and touchscreen smartphones with various internet applications with colorful interfaces, icons, and buttons isolated on a white background.

Johns Hopkins researchers make the case for social media standards on suicide

Johns Hopkins researchers call for the establishment of guidelines that prescribe how modern social media platforms should share pro-social, life-saving education and remove harmful content.

A technology-themed circular graphic projecting a digital brain.

Putting trust to the test

Hopkins researchers unveil new uncertainty quantification methods in an effort to promote appropriate trust in AI use.