An AI model can accurately flag EGFR mutations in lung adenocarcinoma using routine pathology slides, reducing the need for rapid genetic tests.
Key Details
- 1Researchers from Mount Sinai, Memorial Sloan Kettering, and collaborators published results in Nature Medicine on July 9, 2025.
- 2The AI model predicts EGFR mutations from H&E-stained pathology slides of lung adenocarcinoma.
- 3A live 'silent trial' at Memorial Sloan Kettering showed the model could reduce rapid genetic testing by over 40%.
- 4The model was trained and validated on the largest multi-institutional dataset of matched slides and sequencing results from the US and Europe.
- 5Preserving tissue by avoiding unnecessary rapid tests allows for more comprehensive genomic sequencing.
- 6Work is ongoing to broaden the model's biomarker detection and deploy it in more healthcare settings.
Why It Matters
This study demonstrates the practical integration of AI in pathology workflows, expediting precision therapy decisions and optimizing tissue usage in lung cancer diagnostics. It signals a major step toward AI-driven personalization in oncology, with potential to improve both efficiency and patient outcomes.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.