Leveraging Pretrained Neural Network Models for the Classification of Tumor Cells Analyzed by Label-Free Phase Holotomographic Microscopy.

We present an innovative methodology for label-free, high-resolution imaging using phase holotomographic microscopy, coupled with neural network models for the classification of cancer cells. Using 3-dimensional phase holotomographic microscopy, we imaged live A549 lung cancer cells with and without paclitaxel, converted stacks to 2-dimensional maximum-intensity projections, and evaluated pretrained convolutional networks (VGG16, ResNet18, DenseNet121, and EfficientNet-B0) for binary classification of treatment status. EfficientNet-B0 achieved 96.9% accuracy on unsegmented images. Refractive index analysis revealed bimodal distribution in treated cells, reflecting heterogeneous biophysical responses to paclitaxel exposure and supporting the network's ability to detect subtle, label-free indicators of drug action. As further proof of concept, the same pipeline separated holotomographic images of label-free, high- versus low-grade urothelial cancer cells with high accuracy (90.6%). These findings highlight the potential of integrating label-free holotomographic imaging with deep learning techniques for rapid and efficient classification of tumor cells, paving the way for advancements in treatment optimization and personalized diagnostic strategies.