1
|
Martell MT, Haven NJM, Cikaluk BD, Restall BS, McAlister EA, Mittal R, Adam BA, Giannakopoulos N, Peiris L, Silverman S, Deschenes J, Li X, Zemp RJ. Deep learning-enabled realistic virtual histology with ultraviolet photoacoustic remote sensing microscopy. Nat Commun 2023; 14:5967. [PMID: 37749108 PMCID: PMC10519961 DOI: 10.1038/s41467-023-41574-2] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/02/2022] [Accepted: 09/11/2023] [Indexed: 09/27/2023] Open
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.
Collapse
Affiliation(s)
- Matthew T Martell
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Nathaniel J M Haven
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Brendyn D Cikaluk
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Brendon S Restall
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Ewan A McAlister
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Rohan Mittal
- Department of Laboratory Medicine and Pathology, University of Alberta, 11405 87 Avenue NW, Edmonton, AB, T6G 1C9, Canada
| | - Benjamin A Adam
- Department of Laboratory Medicine and Pathology, University of Alberta, 11405 87 Avenue NW, Edmonton, AB, T6G 1C9, Canada
| | - Nadia Giannakopoulos
- Department of Laboratory Medicine and Pathology, University of Alberta, 11405 87 Avenue NW, Edmonton, AB, T6G 1C9, Canada
| | - Lashan Peiris
- Department of Surgery, University of Alberta, 8440 - 112 Street, Edmonton, AB, T6G 2B7, Canada
| | - Sveta Silverman
- Department of Laboratory Medicine and Pathology, University of Alberta, 11405 87 Avenue NW, Edmonton, AB, T6G 1C9, Canada
| | - Jean Deschenes
- Department of Laboratory Medicine and Pathology, University of Alberta, 11405 87 Avenue NW, Edmonton, AB, T6G 1C9, Canada
| | - Xingyu Li
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada
| | - Roger J Zemp
- Department of Electrical and Computer Engineering, University of Alberta, 116 Street & 85 Avenue, Edmonton, AB, T6G 2R3, Canada.
| |
Collapse
|
2
|
Cikaluk BD, Restall BS, Haven NJM, Martell MT, McAlister EA, Zemp RJ. Rapid ultraviolet photoacoustic remote sensing microscopy using voice-coil stage scanning. OPTICS EXPRESS 2023; 31:10136-10149. [PMID: 37157568 DOI: 10.1364/oe.481313] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/10/2023]
Abstract
There is an unmet need for fast virtual histology technologies that exhibit histological realism and can scan large sections of fresh tissue within intraoperative time-frames. Ultraviolet photoacoustic remote sensing microscopy (UV-PARS) is an emerging imaging modality capable of producing virtual histology images that show good concordance to conventional histology stains. However, a UV-PARS scanning system that can perform rapid intraoperative imaging over mm-scale fields-of-view at fine resolution (<500 nm) has yet to be demonstrated. In this work, we present a UV-PARS system which utilizes voice-coil stage scanning to demonstrate finely resolved images for 2×2 mm2 areas at 500 nm sampling resolution in 1.33 minutes and coarsely resolved images for 4×4 mm2 areas at 900 nm sampling resolution in 2.5 minutes. The results of this work demonstrate the speed and resolution capabilities of the UV-PARS voice-coil system and further develop the potential for UV-PARS microscopy to be employed in a clinical setting.
Collapse
|
3
|
Pellegrino N, Ecclestone BR, Dinakaran D, van Landeghem F, Fieguth P, Haji Reza P. Time-domain feature extraction for target specificity in photoacoustic remote sensing microscopy. OPTICS LETTERS 2022; 47:3952-3955. [PMID: 35913356 DOI: 10.1364/ol.457142] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/28/2022] [Accepted: 06/24/2022] [Indexed: 06/15/2023]
Abstract
Photoacoustic remote sensing (PARS) microscopy is an emerging label-free optical absorption imaging modality. PARS operates by capturing nanosecond-scale optical fluctuations produced by photoacoustic pressures. These time-domain (TD) variations are usually projected by amplitude to determine optical absorption magnitude. However, valuable details on a target's material properties (e.g., density, speed of sound) are contained within the TD signals. This work uses a novel, to the best of our knowledge, clustering method to learn TD features, based on signal shape, which relate to underlying material traits. A modified K-means method is used to cluster TD data, capturing representative signal features. These features are then used to form virtual colorizations which may highlight tissues based on their underlying material properties. Applied in fresh resected murine brain tissue, colorized visualizations highlight distinct regions of tissue. This may potentially facilitate differentiation of tissue constituents (e.g., myelinated and unmyelinated axons, cell nuclei) in a single acquisition.
Collapse
|
4
|
Restall BS, Cikaluk BD, Martell MT, Haven NJM, Mittal R, Silverman S, Peiris L, Deschenes J, Adam BA, Kinnaird A, Zemp RJ. Fast hybrid optomechanical scanning photoacoustic remote sensing microscopy for virtual histology. BIOMEDICAL OPTICS EXPRESS 2022; 13:39-47. [PMID: 35154852 PMCID: PMC8803023 DOI: 10.1364/boe.443751] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/20/2021] [Revised: 11/01/2021] [Accepted: 11/23/2021] [Indexed: 05/25/2023]
Abstract
A rapid scanning microscopy method for hematoxylin and eosin (H&E) like images is sought after for interoperative diagnosis of solid tumor margins. The rapid observation and diagnosis of histological samples can greatly lower surgical risk and improve patient outcomes from solid tumor resection surgeries. Photoacoustic remote sensing (PARS) has recently been demonstrated to provide images of virtual H&E stains with excellent concordance with true H&E staining of formalin-fixed, paraffin embedded tissues. By using PARS with constant velocity and 1D galvanometer mirror scanning we acquire large virtual H&E images (10mm x 5mm) of prostate tissue in less than 3.5 minutes without staining, and over two orders of magnitude faster data acquisition than the current PARS imaging speed.
Collapse
Affiliation(s)
- Brendon S. Restall
- University of Alberta, Electrical and Computer Engineering Department, Edmonton, Canada
| | - Brendyn D. Cikaluk
- University of Alberta, Electrical and Computer Engineering Department, Edmonton, Canada
| | - Matthew. T. Martell
- University of Alberta, Electrical and Computer Engineering Department, Edmonton, Canada
| | - Nathaniel J. M. Haven
- University of Alberta, Electrical and Computer Engineering Department, Edmonton, Canada
| | - Rohan Mittal
- Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Canada
| | - Sveta Silverman
- Laboratory Medicine, Misericordia Hospital, Edmonton, Canada
| | - Lashan Peiris
- Division of Department of Surgery, University of Alberta, Edmonton, Canada
| | - Jean Deschenes
- Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Canada
| | - Benjamin A. Adam
- Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Canada
| | - Adam Kinnaird
- Division of Urology, Department of Surgery, University of Alberta, Edmonton, Canada
| | - Roger J. Zemp
- University of Alberta, Electrical and Computer Engineering Department, Edmonton, Canada
| |
Collapse
|
5
|
Haven NJM, Martell MT, Cikaluk BD, Restall BS, McAlister E, Silverman S, Peiris L, Deschenes J, Li X, Zemp RJ. Virtual histopathology with ultraviolet scattering and photoacoustic remote sensing microscopy. OPTICS LETTERS 2021; 46:5153-5156. [PMID: 34653139 DOI: 10.1364/ol.436136] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/09/2021] [Accepted: 08/19/2021] [Indexed: 06/13/2023]
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.
Collapse
|