Label-free imaging of collagen fibers in tissue slices using phase imaging with computational specificity

Evaluating the tissue collagen content in addition to the epithelial morphology has been proven to offer complementary information in histopathology, especially in oncology tumor staging and prediction of survival in cancer patients. One imaging modality widely used for this purpose is second harmonic generation microscopy (SHGM), which reports on the nonlinear susceptibility associated with the collagen fibers. Another method is polarization light microscopy (PLM) combined with picrosirius-red (PSR) tissue staining. However, SHGM requires expensive equipment and provides limited throughput, while PLM and PSR staining are not part of the routine surgical pathology workflow. Here, we utilize phase imaging with computational specificity (PICS) to computationally infer the collagen distribution of unlabeled tissue, with high specificity. PICS utilizes deep learning to translate quantitative phase images (QPI) into corresponding PSR images with high accuracy and inference speed of 200 milisecond per forwardpass through the model once trained. We developed a multimodal imaging instrument that yields both Spatial light Inference Microscopy (SLIM) and polarized light microscopy (PLM) images from the same field of view. Our results indicate that the distributions of collagen fiber orientation, length, and straightness reported by PICS closely match the ones from ground truth as defined by KL-divergence.