AI-Enhanced Microfluidic Chip Enables Label-Free Detection of Circulating Tumor Cells

Researchers developed a microfluidic chip and AI system that accurately isolates and identifies rare circulating tumor cells (CTCs) from blood without fluorescent labeling.
Key Details
- 1The platform uses contraction-expansion inertial microfluidics to separate CTCs based on size differences from blood cells.
- 2Identification of enriched tumor cells is performed with the YOLOv8 deep learning model using standard bright-field microscope images.
- 3The system achieved an MCF-7 cell recovery rate of 89.2±3.1% and a white blood cell removal rate of 86.9±1.4%.
- 4YOLOv8 attained 96% accuracy, 94.6% precision and recall, and 95.4% specificity in recognizing tumor cells.
- 5The method eliminates the need for cell-surface marker labeling or fluorescent staining.
- 6Validation was conducted on artificial blood samples, with future studies planned for patient samples.
Why It Matters

Source
EurekAlert
Related News

AI Heart Age Estimate from Ultrasound Predicts Long-Term Cardiovascular and Dementia Risk
AI-derived heart age from echocardiography images exceeds actual age: linked with higher cardiovascular and dementia mortality over 20 years.

Machine Learning Models Improve Mortality Prediction Post-Heart Attack
New machine-learning models outperform traditional risk scores for predicting mortality after a heart attack.

MRI Study Reveals Disease Epicenters Driving Atrophy in Multiple Sclerosis
Large-scale MRI analysis identifies specific brain regions acting as epicenters of atrophy in multiple sclerosis patients.