Virtual histopathology with ultraviolet scattering and photoacoustic remote sensing microscopy

Abstract

Realistic label-free virtual histopathology has been a long sought-after goal not yet achieved with current methods. Here, we introduce high-resolution hematoxylin and eosin (H&E)-like virtual histology of unstained human breast lumpectomy specimen sections using ultraviolet scattering-augmented photoacoustic remote sensing microscopy. Together with a colormap-matching algorithm based on blind stain separation from a reference true H&E image, we are able to produce virtual H&E images of unstained tissues with close concordance to true H&E-stained sections, with promising diagnostic utility.

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

“…Methods which require staining or optical clearing procedures usually require an additional few minutes which can be eliminated in our label-free approach. Our rapid 7 mins/cm 2 acquisition time is significantly faster than scan times reported for SRS microscopy 35 , 36 , recent UV-PAM imaging systems 43 , 46 , 48 , multispectral deep-UV microscopy 37 or previous UV-PARS work 12 – 14 , 20 , while maintaining a high 390 nm resolution. Overall, our system offers a well-positioned trade-off of scan speed and resolution, while providing label-free contrasts that correspond well to gold-standard H&E-stained histology.…”

“…Integration of near-infrared (NIR) scattering microscopy further enabled simultaneous acquisition of complementary virtual eosin contrast co-registered with the UV-PARS images 15 – 17 , simplifying the complexity of previous dual-contrast approaches 18 , 19 . The NIR scattering was later upgraded to 266 nm pulsed UV scattering along with a pulse peak sample-and-hold detection circuit to achieve improved resolution in the virtual eosin data 20 utilizing the existing excitation laser source. In this work we utilize a high-resolution 0.5 numerical aperture (NA) objective with rapid voice coil scanning capable of imaging at speeds of up to 7 mins/cm 2 with 390 nm lateral resolution 21 , and axial optical sectioning of 1.6 μm similar to thin histology sections 22 .…”

Ultraviolet photoacoustic remote sensing and scattering microscopy for CycleGAN-enabled realistic virtual histology

“…This method is easily integrated with 266 nm optical scattering microscopy with contrast resembling an eosin stain. 3 Images in unstained tissues using this dual-channel system can be pseudo-colored and combined for a complete virtual histological visualization, with features that are concordant with conventional H&E staining. However, a disparity in stain style remains between virtual histology and the gold standard.…”

Investigating mechanisms of laser pulse-induced reflectivity modulations in photoacoustic remote sensing with a 10 million frames-per-second camera

“…3g may also be emitted material following ablation of the fiber. Next, to explore potential laser-induced sample motion in liquid media, 25 µm diameter gold wires were submerged in a thin layer of distilled water and covered with a coverslip to follow previously reported sample preparation used in PARS microscopy experiments 10 . In the presence of water, laser-induced bubble formation was observed with bubble dissipation occurring as early as 10 µs post-excitation.…”

“…Photoacoustic microscopy (PAM) generates absorption-contrast imagery by tuning an excitation wavelength to match chromophore absorption peaks and using an ultrasound transducer to detect acoustic waves generated via the photoacoustic effect following excitation pulse absorption 1 . Recently, we developed photoacoustic remote sensing (PARS) microscopy, an all-optical non-contact version of PAM, demonstrating high sensitivity to a wide variety of endogenous chromophores [2][3][4][5] and ready integration into existing technologies due to its all-optical architecture [6][7][8][9][10] . In comparison to PAM, PARS microscopy utilizes an additional interrogation beam to probe reflectivity modulations rather than directly detecting generated acoustic waves using an ultrasound transducer.…”