Parsin Haji Reza

James E. D. Tweel, Benjamin R. Ecclestone, Marian Boktor et al.

The field of histology relies heavily on antiquated tissue processing and staining techniques that limit the efficiency of pathologic diagnoses of cancer and other diseases. Current staining and advanced labeling methods are often destructive and mutually incompatible, requiring new tissue sections for each stain. This prolongs the diagnostic process and depletes valuable biopsy samples. In this study, we present an alternative label-free histology platform using the first transmission-mode Photon Absorption Remote Sensing microscope. Optimized for automated whole slide scanning of unstained tissue samples, the system provides slide images at magnifications up to 40x that are fully compatible with existing digital pathology tools. The scans capture high quality and high-resolution images with subcellular diagnostic detail. After imaging, samples remain suitable for histochemical, immunohistochemical, and other staining techniques. Scattering and absorption (radiative and non-radiative) contrasts are shown in whole slide images of malignant human breast and skin tissues samples. Clinically relevant features are highlighted, and close correspondence and analogous contrast is demonstrated with one-to-one gold standard H&E stained images. Our previously reported pix2pix virtual staining model is applied to an entire whole slide image, showcasing the potential of this approach in whole slide label-free H&E emulation. This work is a critical advance for integrating label-free optical methods into standard histopathology workflows, both enhancing diagnostic efficiency, and broadening the number of stains that can be applied while preserving valuable tissue samples.

Qiankai Wang, James E. D. Tweel, Parsin Haji Reza et al.

Virtual staining has emerged as a powerful alternative to traditional histopathological staining techniques, enabling rapid, reagent-free image transformations. However, existing evaluation methods predominantly rely on full-reference image quality assessment (FR-IQA) metrics such as structural similarity, which are originally designed for natural images and often fail to capture pathology-relevant features. Expert pathology reviews have also been used, but they are inherently subjective and time-consuming. In this study, we introduce PaPIS (Pathology-Aware Perceptual Image Similarity), a novel FR-IQA metric specifically tailored for virtual staining evaluation. PaPIS leverages deep learning-based features trained on cell morphology segmentation and incorporates Retinex-inspired feature decomposition to better reflect histological perceptual quality. Comparative experiments demonstrate that PaPIS more accurately aligns with pathology-relevant visual cues and distinguishes subtle cellular structures that traditional and existing perceptual metrics tend to overlook. Furthermore, integrating PaPIS as a guiding loss function in a virtual staining model leads to improved histological fidelity. This work highlights the critical need for pathology-aware evaluation frameworks to advance the development and clinical readiness of virtual staining technologies.

Nicholas Pellegrino, Paul Fieguth, Parsin Haji Reza

Many measurement modalities which perform imaging by probing an object pixel-by-pixel, such as via Photoacoustic Microscopy, produce a multi-dimensional feature (typically a time-domain signal) at each pixel. In principle, the many degrees of freedom in the time-domain signal would admit the possibility of significant multi-modal information being implicitly present, much more than a single scalar "brightness", regarding the underlying targets being observed. However, the measured signal is neither a weighted-sum of basis functions (such as principal components) nor one of a set of prototypes (K-means), which has motivated the novel clustering method proposed here. Signals are clustered based on their shape, but not amplitude, via angular distance and centroids are calculated as the direction of maximal intra-cluster variance, resulting in a clustering algorithm capable of learning centroids (signal shapes) that are related to the underlying, albeit unknown, target characteristics in a scalable and noise-robust manner.