Deep learning-enabled realistic virtual histology with ultraviolet photoacoustic remote sensing microscopy

Abstract

The goal of oncologic surgeries is complete tumor resection, yet positive margins are frequently found postoperatively using gold standard H&E-stained histology methods. Frozen section analysis is sometimes performed for rapid intraoperative margin evaluation, albeit with known inaccuracies. Here, we introduce a label-free histological imaging method based on an ultraviolet photoacoustic remote sensing and scattering microscope, combined with unsupervised deep learning using a cycle-consistent generative adversarial network for realistic virtual staining. Unstained tissues are scanned at rates of up to 7 mins/cm2, at resolution equivalent to 400x digital histopathology. Quantitative validation suggests strong concordance with conventional histology in benign and malignant prostate and breast tissues. In diagnostic utility studies we demonstrate a mean sensitivity and specificity of 0.96 and 0.91 in breast specimens, and respectively 0.87 and 0.94 in prostate specimens. We also find virtual stain quality is preferred (P = 0.03) compared to frozen section analysis in a blinded survey of pathologists.

Automated Whole Slide Imaging for Label-Free Histology Using Photon Absorption Remote Sensing Microscopy

“…Specifically, in this configuration, the detection power incident on the sample during an acquisition was in the range of ~2-5µW compared to the previously reported value of 156µW [56]. In addition, the pulse energies from UV source were in the range of 150-200pJ compared to the 5nJ excitation energies used in the most recent UV-PARS histology system [57], [58]. These pulse energies and detection powers permit standard H&E staining and clinical diagnosis following PARS imaging without any visual damage to the sample.…”

“…Additionally, the multi-scale structural similarity index measure (MS-SSIM) was employed as a pixel-level quantitative measure of agreement. Following the methodology in [58], RGB images were converted to the YCbCr color space, and the MS-SSIM metric was computed for each channel and summed together with the following weights: 0.8 (Y), 0.1 (Cb), and 0.1 (Cr). This process was performed for the native PARS 40x magnification and was repeated for 20x, 10x, and 5x equivalent viewing magnifications.…”

“…These optics refinements facilitate histology-like label-free whole slide imaging, and importantly, allow for a significant reduction in detection powers and ultraviolet excitation energy compared to recent PARS histology implementations [57], [58]. With these reduced powers, the unstained tissue sections remain unaltered and available for subsequent standard histochemical staining, enabling us to showcase both PARS and H&E whole slide images of the same tissue section in malignant human breast and skin samples.…”

Photoacoustic imaging for cutaneous melanoma assessment: a comprehensive review

“…Recently, all-optical PA microscopy in reflection-mode has been shown to be able to form histology-like images of various cancers on unstained slides to distinguish tissue types. 142,267 The principle behind this technology is that by concurrently measuring radiative and nonradiative (in the form of acoustic) emissions from tissue samples illuminated with light in the UV range (266 nm), it is possible to differentiate DNA, RNA, collagen, and elastin, among other chromophores. Nuclear contrast, in particular, comes from relaxation of DNA, and nonnuclear contrast from relation of extranuclear proteins.…”

Accelerated Ultraviolet Photoacoustic Microscopy Based on Optical Ultrasound Detection for Breast-Cancer Biopsy

“…This imaging scheme, which can be viewed as an integration of PAM and computed tomography, reduces the scanning time at the expense of complex optical illumination and ultrasound detection configurations, as well as a high-performance, costly data acquisition system. The ultraviolet photoacoustic remote sensing microscopy (UV-PARS) is to measure the PA signals by an all-optical approach [17], [18]. This optical-based UV-PARS can eliminate the needs of acoustic coupling and only need to scanning the optical beam.…”