Public data for nonlinear optical computational histology

Experimental datasets for training and testing nonlinear optical computational histology (NOCH), which include label-free nonlinear optical data: stimulated Raman scattering (SRS) images of human brain tumors and multiphoton (MP) images of human ovarian cancers, and the corresponding H&E slices. The contrastive deep learning framework can generate diagnostic quality H&E slides comparable to conventional histopathology. If you find this work useful in your research, please consider citing the paper: B. Shen, Z. Li, Y. Pan, Y. Guo, Z. Yin, R. Hu, J. Qu, L. Liu, Noninvasive Nonlinear Optical Computational Histology. Adv. Sci. 2023, 2308630.