We demonstrate a rapid deep ultraviolet transmission and scattering microscopy system which utilizes a single 266 nm nanosecond pulsed laser source to generate virtual histology images. By detecting both the transmitted and back-scattered 266 nm radiation, distinct contrasts of the tissue section are produced. Maximally realistic virtual histology images are then generated using each contrast as separate inputs to a CycleGAN deep neural network. This work demonstrates virtual histology images of mm-scale areas at fine resolutions (< 500 nm) within minutes. This system greatly simplifies the optics and alignment required for imaging, and could provide future utility for clinical studies.
Ultraviolet photacoustic remote sensing (UV-PARS) microscopy is a non-contact, label-free imaging modality which has demonstrated the ability to generate virtual histology images that show good concordance to gold-standard hematoxylin and eosin (H&E) stains of fixed tissue sections. However, UV-PARS microscopy requires time-consuming and complex coalignment of multiple beams for imaging. In this work, we demonstrate UV transmittance and scattering microscopy for virtual histology and compare with UV-PARS microscopy. Maximally realistic virtual histology images are generated for both UV-PARS and UV-transmittance microscopy techniques using a cycle-consistent generative adversarial network (CycleGAN) and compared to one another alongside the gold-standard brightfield microscopy of H&E stained fixed tissues.
Pathologist evaluation of brightfield microscopy images of H&E-stained tissues is time-consuming and labor-intensive, preventing intraoperative feedback to surgeons. Combined ultraviolet reflectance and photoacoustic remote sensing (PARS) microscopy was recently introduced, providing a label-free way to obtain cell nuclei and cytoplasm images. Additionally, utilizing deep learning and fast voice-coil scanning, maximally realistic virtual histology can be rapidly generated with this technique. Here, we introduce combined ultraviolet confocal reflectance, autofluorescence, and PARS microscopy, utilizing a single 266 nm excitation beam. This system now allows us to generate sectioned virtual histology in thick tissues well suited to the intraoperative setting.
Ultraviolet photoacoustic remote sensing microscopy provides label-free optical absorption contrast comparable to hematoxylin staining. This has been combined with 266 nm optical scattering microscopy offering eosin-like contrast. Here, we use unsupervised deep learning-based style transfer using the CycleGAN approach to render these pseudo-colored virtual histological images in a realistic stain style comparable to the H&E gold standard in unstained human and murine tissue specimens. A multi-pathologist diagnostic concordance study found a sensitivity of 89%, specificity of 91%, and accuracy of 90%. A blinded subjective stain quality survey suggested virtual histology was preferred over frozen sections at the 95% confidence level.
We present photoacoustic remote sensing and autofluorescence microscopy with surface excitation (PARS-AMUSE) system that combines several methodologies of optical imaging including absorption, scattering and radiative emission. We generate virtual hematoxylin and eosin (H&E) images from absorption and scattering data using a generative adversarial network for margin assessment and cancer grading with simultaneous metabolic measurements from the autofluorescence of NADH and FAD. This system utilizes a single excitation source for all three imaging modalities to minimize complexity and cost. Using this system, we aim to profile specific cancers and predict cancer aggression.
We present a novel microscopy technique called photoacoustic remote sensing combined with autofluorescence with surface excitation (PARS-AMUSE) to generate virtual hematoxylin and eosin (H&E) images with simultaneous metabolic measurements to predict cancer aggression. PARS uses absorption and scattering contrast for the virtual H&E images to visualize structural information, perform margin assessment and grade cancers. AMUSE utilizes the same light source used in PARS to excite NADH and FAD autofluorescence for quantifying the optical redox ratio (ORR), an important indicator of tissue metabolism. Using our system, we aim to differentiate receptor status and predict cancer aggression.
We demonstrate a rapid photoacoustic remote sensing (PARS) microscopy system with voice-coil stage scanning, capable of acquiring 1cm2 virtual histology images at 800nm x 800nm lateral sampling density in 3.75 minutes, high-resolution (300nm x 300nm sampling density) images at 1mm2 in 20 seconds and 300μm x 300μm in 3 seconds. The combined resolution and scanning speed that can be achieved by this system may pave way for clinical translation to rapidly assess resected tumor margins in the surgical suite.
Previous photoacoustic remote sensing virtual histology approaches were too slow to use intraoperatively. We present a new scanning methodology with simultaneous galvanometer mirror and constant velocity mechanical scanning to greatly increase image acquisition speed. Human breast and prostate samples are imaged over an area of 4mm x 4mm in 40s with a 0.5μm resolution resolving both cancerous and healthy tissue. Histological detail is clearly visible in our images where tissue organization and subcellular nuclei density can be observed to aid histologists in determining margin status and cancer grading.
Currently, there is an inability to obtain fast realistic label-free virtual histopathological images of tissues. We previously introduced ultraviolet photoacoustic remote sensing microscopy as a method to obtain virtual hematoxylin contrast albeit without the ability to obtain virtual eosin contrast. By utilizing UV scattering as a high-resolution eosin channel we are able to produce complete H&E-like virtual histology of unstained human breast lumpectomy specimen sections. By further leveraging a novel colormap matching algorithm with this UV scattering, we generate H&E-like output that is shown to have strong concordance with true H&E-stained adjacent sections, showing promising diagnostic utility.
An estimated 20-40% of patients who undergo breast-conserving surgery require repeat operations due to the length of time required post-surgery to analyze the resected sample. Recently, a novel non-contact microscopy technique known as ultraviolet photoacoustic remote sensing (UV-PARS) has been developed, which can produce virtual hematoxylin and eosin (H&E) stained images. By using a voice-coil scanning stage in conjunction with an on-demand pulsed laser, we demonstrate 1cm x 1cm gross UV-PARS virtual H&E scans in approximately 50 seconds, along with full-resolution 1cm x 1cm scans in approximately 8 minutes. With widefield high-speed scanning, UV-PARS shows promise for future translation to clinical application.
We demonstrate the ability to perform fast ultraviolet photoacoustic remote sensing microscopy of tissue margins using a mosaic scanning approach where a camera-based system is used to determine scan coordinates within a 2-mm margin.
We demonstrate the use of Photoacoustic Remote Sensing (PARS) and scattering microscopy capable of acquiring virtual depth-resolved images of tissues with virtual contrast of hematoxylin using PARS & eosin using scattering microscopy.
Hematoxylin and Eosin (H and E) staining is the gold standard for the majority of histopathological diagnostics but requires lengthy processing times not suitable for point-of-care diagnosis. Here we demonstrate a 266-nm excitation Ultraviolet Photoacoustic Remote Sensing (UV-PARS) and Scattering Microscopy system capable of virtual H and E 3D imaging of tissues in conjunction with with confocal fluorescence microscopy (CFM) for validation in thick tissues. We demonstrate the capabilities of this dual-contrast system for en-face planar and volumetric imaging of human tissue samples exhibiting high concordance with the gold standard of H and E staining procedures as well as confocal fluorescence microscopy. To our knowledge, this is the first near real-time microscopy approach capable of volumetric imaging unstained thick tissues with virtual H and E contrast.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.