A new metasurface spectral AI chip enables rapid, accurate diagnosis of meibomian gland dysfunction (MGD) from tissue samples, achieving 96.22% accuracy.
Key Details
- 1Researchers from Peking Union Medical College Hospital and Tsinghua University developed a compact optical AI chip for MGD diagnosis.
- 2The AI chip integrates a spectral convolutional neural network (SCNN) directly onto a CMOS imaging sensor.
- 3Pathological tissue sections from MGD patients and controls were rapidly analyzed, producing spectral feature maps in tens of milliseconds.
- 4The chip identifies MGD with 96.22% accuracy, outperforming conventional RGB image analysis (84%) and matching hyperspectral imaging.
- 5Distinct spectral signatures corresponded with disease severity and biochemical changes in gland tissue.
- 6The device is CMOS-fabricated, enabling potential clinical integration for real-time, objective ocular assessment.
Why It Matters
This innovation demonstrates the power of combining metasurface optics and AI for rapid, objective tissue diagnostics—reducing reliance on subjective clinical assessment and paving the way for real-time, in-sensor medical AI applications.
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.