Researchers merged survival analysis with machine learning on 124,901 personal health records to create individualized predictions of healthy years remaining—a significant step toward precision aging medicine. Original paper: Bridging survival analysis and machine learning to improve healthy life expectancy estimation using PHR records. — NPJ digital medicine. 10.1038/s41746-026-02700-8 📄 Read the article

Automated liver segmentation using deep learning accelerates preoperative assessment for hepatic resection, reducing analysis time 8-fold while maintaining high diagnostic accuracy. Original paper: Automated Delineation of Couinaud Segments on CT for Future Liver Remnant Volumetry. — Radiology. Artificial intelligence. 10.1148/ryai.260506 📄 Read the article

Researchers developed a machine learning algorithm that generates synthetic CT images from MRI scans of the temporal bone, potentially eliminating the need for separate radiation-based imaging in surgical planning. Original paper: Machine Learning-Based Synthetic Computed Tomography Derived From Temporal Bone Magnetic Resonance Imaging. — JAMA otolaryngology– head & neck surgery. 10.1001/jamaoto.2026.0432 📄 Read the article

A multicenter study of 3,031 emergency CT scans shows AIDOC-VO, the first commercial AI tool for detecting intracranial vessel occlusions, performs comparably to expert radiologists while identifying clinically significant missed cases. Original paper: Commercial AI Model Diagnostic Accuracy for Intracranial Large- and Medium-Vessel Occlusion in Emergency CT Angiography. — Radiology. Artificial intelligence. 10.1148/ryai.250749 📄 Read the article

A large language model fine-tuned on 1.87 million radiology reports generates clinically equivalent impressions while reducing reporting time by nearly half, according to a multicenter evaluation across 42 Chinese hospitals. Original paper: Fine-Tuned Large Language Models for Automated Radiology Impression Generation: A Multicenter Evaluation. — Radiology. Artificial intelligence. 10.1148/ryai.250714 📄 Read the article

A randomized trial demonstrates that AI-assisted dermatology tools substantially improve how accurately patients name and understand their skin conditions. Original paper: Consumer Understanding of Skin Concerns With an AI-Powered Informational Tool. — JAMA Dermatology. 10.1001/jamadermatol.2026.0597 📄 Read the article

Researchers discovered that large language models tested on radiology board-style questions are highly vulnerable to cognitively biased prompts, with accuracy plummeting by up to 45% under certain attack conditions. Original paper: Cognitively Biased Prompt Effects on Large Language Model Accuracy for Radiology Board-Style Examination Questions. — Radiology. Artificial intelligence. 10.1148/ryai.250585 📄 Read the article

Researchers have synthesized evidence on AI implementation barriers into FAIIR-H, a comprehensive framework that reveals why general implementation science approaches fall short for AI technologies in clinical settings. Original paper: Framework for artificial intelligence implementation research in healthcare: synthesizing current evidence on barriers and facilitators. — NPJ digital medicine. 10.1038/s41746-026-02705-3 📄 Read the article

A new lightweight AI system called Janus-Pro-CXR, powered by DeepSeek’s multimodal model, significantly improves chest radiograph interpretation quality and efficiency in clinical practice. Original paper: A DeepSeek-powered AI system for automated chest radiograph interpretation in clinical practice. — Nature communications. 10.1038/s41467-026-72680-6 📄 Read the article

A new smartphone-compatible AI system accurately screens for multiple anterior segment diseases in real time, offering a scalable solution for early detection in underserved communities. Original paper: Smartphone-based lightweight AI system for real-time multiple anterior segment disease screening: development and real-world validation. — BMC medicine. 10.1186/s12916-026-04905-w 📄 Read the article