Abstract: Described herein are techniques for denoising images of samples produced using polarization microscopy. Polarization images captured by polarization microscopy contain not only signals produced from polarization, but also signals produced from scattering by artifacts (e.g., crystals and/or hemosiderin). These off-target signals reduce the signal-to-noise ratio (SNR) of the birefringent substance of interest, and reduce the utility and accuracy of polarization microscopy for substance quantification. The techniques developed by the inventors and described herein reduce noise resulting from scattering, thus enabling much cleaner, noise-reduced images. The images so produced can be either directly visualized by a medical practitioner or used for downstream machine-learning models. Noise reduction involves i) dividing an image in segments (e.g., pixels or groups of pixels), ii) performing spectral analysis of each segment, and iii) separating each segment on the basis of its spectral profiles.

Abstract: In some aspects, a method, a system, or a non-transitory computer-readable storage medium are described for using a machine learning (ML) model to obtain annotations of a pathology slide image obtained in a first imaging modality, where the ML model is trained based in part on images obtained from a second imaging modality different from the first imaging modality. The first imaging modality is a conventional scanner for whole-slide images (WSI). The second imaging modality may include one or more of multispectral imaging (MSI), polarization imaging, quantitative phase imaging, or a combination thereof. The trained ML model can generate annotations that include more details with higher accuracy in comparison to annotating based on the WSI images alone.