Researchers have developed RBT-GC, a blood-based biomarker that identifies high-risk gastric cancer patients among those with chronic dyspepsia, dramatically outperforming traditional tumor markers like CEA and CA19.9. Original paper: Development and validation of a novel blood-based biomarker for gastric cancer triage in chronic dyspepsia. — NPJ digital medicine. 10.1038/s41746-026-02618-1 📄 Read the article

Researchers have developed an interpretable AI system that combines baseline PET/CT imaging with electronic health records to predict progression-free survival and overall survival in mantle cell lymphoma patients with unprecedented accuracy. Original paper: Interpretable multimodal PET/CT-EHR fusion via mixture-of-experts for prognostic stratification in mantle cell lymphoma: a multicenter study. — BMC medicine. 10.1186/s12916-026-04865-1 📄 Read the article

A multi-center study demonstrates that combining PET/CT and MRI with machine learning significantly improves detection of clinically significant prostate cancers. Original paper: Non-invasive diagnosis strategy integrating PSMA PET/CT and mpMRI for patients with suspected prostate cancer: a multi-center study. — BMC medicine. 10.1186/s12916-026-04859-z 📄 Read the article

A new automated deep learning system can detect focal liver lesions in ultrasound videos in real-time, and critically, it levels the playing field—helping less experienced sonographers match the performance of veteran specialists. Original paper: Automated deep learning for real-time focal liver lesions detection in ultrasound videos a multicenter study. — NPJ digital medicine. 10.1038/s41746-026-02629-y 📄 Read the article

A novel deep learning model combining digital pathology with clinical data accurately predicts early recurrence in gastric cancer, while identifying a subset of patients with immune-activated tumors potentially responsive to immunotherapy. Original paper: A deep learning-based digital biopsy for predicting early recurrence in gastric cancer. — Nature communications. 10.1038/s41467-026-71347-6 📄 Read the article

A comprehensive framework identifies seven critical failure modes in AI deployment across clinical medicine and proposes new competencies for clinicians and developers to ensure trustworthy implementation. Original paper: Seven deadly sins in artificial intelligence for digital medicine. — NPJ digital medicine. 10.1038/s41746-026-00391-1 📄 Read the article