Radiology AI makes consistent diagnoses using 3D images from different health centres

2026-03-04 · Source: Machine learning : nature.com subject feeds · Field: Health & Wellbeing — Medical Devices & Health Technology, Clinical Care & Medical Practice, Medical Specialties & Subspecialties · Depth: Intermediate, quick

Summary

A 3D vision–language model named Merlin has been developed to meet the increasing demand for radiology artificial-intelligence tools, trained on abdominal computed-tomography scans, radiology reports, and electronic health records. Merlin demonstrated robust "off-the-shelf" performance across three hospital sites distinct from its initial training center. This strong performance highlights Merlin's significant potential for broader clinical adoption in making consistent diagnoses using 3D medical images.

Key takeaway

A new 3D vision-language model, Merlin, achieves consistent radiology diagnoses from diverse health centers. Trained on abdominal CT scans, reports, and EHRs, it demonstrated stronger off-the-shelf performance than other models. This robust generalizability across three distinct hospital sites enables broader clinical adoption for AI/ML professionals developing medical imaging solutions.

Topics

• Radiology AI
• Vision-Language Models
• Computed Tomography
• Medical Imaging
• Clinical Diagnostics

Best for: Computer Vision Engineer, AI Scientist, Research Scientist, AI Researcher, AI Engineer, AI Data Scientist

Related on AIssential

• Benchmarks and methods for 3D medical image retrieval — The 3D-MIR benchmark and methods advance medical image retrieval by providing standardized evaluation and multi-modal search strategies.
• AI shows its skills in the emergency room — An older OpenAI model surpassed human doctors in diagnosing emergency room cases using only text data.
• OpenMedReason: Scientific Reasoning Supervision for Medical Vision-Language Models — Medical LVLMs require reasoning grounded in visual evidence and clinical knowledge, not just correct answers.
• Dual-Modal Lung Cancer AI: Interpretable Radiology and Microscopy with Clinical Risk Integration — Integrating CT radiology and H&E microscopy with clinical data significantly enhances lung cancer diagnosis and interpretability.
• Learning neuroimaging models from health system-scale data — Health system-scale data enables AI foundation models to achieve superior, fair, and explainable neuroimaging diagnostics.
• MetaDent: Labeling Clinical Images for Vision-Language Models in Dentistry — MetaDent provides a novel dataset and benchmarks to advance VLM capabilities in fine-grained intraoral image analysis.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.