MIT and Microsoft researchers created an AI model to design peptide-based sensors for ultra-early cancer detection by detecting cancer-specific enzymes.
Key Details
- 1The AI model, CleaveNet, rapidly designs peptides that are efficiently and specifically cleaved by cancer-linked proteases.
- 2Nanoparticles coated with these peptides serve as in vivo sensors, releasing detectable signals in urine if target proteases are present.
- 3This approach allows non-invasive at-home cancer screening, potentially detecting and distinguishing between up to 30 cancer types.
- 4The AI-enabled sensors demonstrated high specificity for proteases such as MMP13, linked to cancer metastasis.
- 5The research appears in Nature Communications (DOI: 10.1038/s41467-025-67226-1) and is supported by ARPA-H and several foundations.
Why It Matters
AI-guided molecular sensing represents a major step towards ultra-early, highly specific cancer detection, expanding the toolkit for noninvasive diagnostics. This molecular AI approach may complement or ultimately integrate with radiology and imaging AI, enhancing multidisciplinary cancer care and research in biomarkers.
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.