Journal of Biophotonics, Journal Year: 2025, Volume and Issue: unknown
Published: May 20, 2025
ABSTRACT The lack of high‐throughput, label‐free, and intelligent recognition models for assessing cell death hinders the broad application analysis in chemotherapy lung cancer. We propose an quantitative detection technique deaths. Using high‐throughput phase imaging flow cytometry to capture numerous label‐free images employing convolutional neural networks (CNN) characterize heterogeneity death. revealed through morphology features achieved interpretability CNN using clustering. Finally, classification reliability was validated by extracting from classified cells. This method, compared with biochemical methods, showed a correlation 0.92 0.91 autophagy (Pearson Cosine Similarity), average error 12.52% apoptosis detection. Our approach has potential become valuable tool studying mechanisms offers new perspective cancer treatment.
Language: Английский