Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining
Histological images can reveal rich cellular information of tissue sections, which are widely used by pathologists in disease diagnosis. However, the gold standard for histopathological examination is based on thin sections on slides, which involves inevitable time-consuming and labor-intensive tiss...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | Photoacoustics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2213597921000689 |
| _version_ | 1818986631917469696 |
|---|---|
| author | Lei Kang Xiufeng Li Yan Zhang Terence T.W. Wong |
| author_facet | Lei Kang Xiufeng Li Yan Zhang Terence T.W. Wong |
| author_sort | Lei Kang |
| collection | DOAJ |
| description | Histological images can reveal rich cellular information of tissue sections, which are widely used by pathologists in disease diagnosis. However, the gold standard for histopathological examination is based on thin sections on slides, which involves inevitable time-consuming and labor-intensive tissue processing steps, hindering the possibility of intraoperative pathological assessment of the precious patient specimens. Here, by incorporating ultraviolet photoacoustic microscopy (UV-PAM) with deep learning, we show a rapid and label-free histological imaging method that can generate virtually stained histological images (termed Deep-PAM) for both thin sections and thick fresh tissue specimens. With the tissue non-destructive nature of UV-PAM, the imaged intact specimens can be reused for other ancillary tests. We demonstrated Deep-PAM on various tissue preparation protocols, including formalin-fixation and paraffin-embedding sections (7-µm thick) and frozen sections (7-µm thick) in traditional histology, and rapid assessment of intact fresh tissue (~ 2-mm thick, within 15 min for a tissue with a surface area of 5 mm × 5 mm). Deep-PAM potentially serves as a comprehensive histological imaging method that can be simultaneously applied in preoperative, intraoperative, and postoperative disease diagnosis. |
| first_indexed | 2024-12-20T18:53:52Z |
| format | Article |
| id | doaj.art-f1a28929bbc147a085ea06e694a39a5c |
| institution | Directory Open Access Journal |
| issn | 2213-5979 |
| language | English |
| last_indexed | 2024-12-20T18:53:52Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Photoacoustics |
| spelling | doaj.art-f1a28929bbc147a085ea06e694a39a5c2022-12-21T19:29:34ZengElsevierPhotoacoustics2213-59792022-03-0125100308Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual stainingLei Kang0Xiufeng Li1Yan Zhang2Terence T.W. Wong3Translational and Advanced Bioimaging Laboratory, Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong, ChinaTranslational and Advanced Bioimaging Laboratory, Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong, ChinaTranslational and Advanced Bioimaging Laboratory, Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong, ChinaCorresponding author.; Translational and Advanced Bioimaging Laboratory, Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong, ChinaHistological images can reveal rich cellular information of tissue sections, which are widely used by pathologists in disease diagnosis. However, the gold standard for histopathological examination is based on thin sections on slides, which involves inevitable time-consuming and labor-intensive tissue processing steps, hindering the possibility of intraoperative pathological assessment of the precious patient specimens. Here, by incorporating ultraviolet photoacoustic microscopy (UV-PAM) with deep learning, we show a rapid and label-free histological imaging method that can generate virtually stained histological images (termed Deep-PAM) for both thin sections and thick fresh tissue specimens. With the tissue non-destructive nature of UV-PAM, the imaged intact specimens can be reused for other ancillary tests. We demonstrated Deep-PAM on various tissue preparation protocols, including formalin-fixation and paraffin-embedding sections (7-µm thick) and frozen sections (7-µm thick) in traditional histology, and rapid assessment of intact fresh tissue (~ 2-mm thick, within 15 min for a tissue with a surface area of 5 mm × 5 mm). Deep-PAM potentially serves as a comprehensive histological imaging method that can be simultaneously applied in preoperative, intraoperative, and postoperative disease diagnosis.http://www.sciencedirect.com/science/article/pii/S2213597921000689Deep learningUnsupervised learningPhotoacoustic microscopyHistological imaging |
| spellingShingle | Lei Kang Xiufeng Li Yan Zhang Terence T.W. Wong Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining Photoacoustics Deep learning Unsupervised learning Photoacoustic microscopy Histological imaging |
| title | Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining |
| title_full | Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining |
| title_fullStr | Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining |
| title_full_unstemmed | Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining |
| title_short | Deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real-time virtual staining |
| title_sort | deep learning enables ultraviolet photoacoustic microscopy based histological imaging with near real time virtual staining |
| topic | Deep learning Unsupervised learning Photoacoustic microscopy Histological imaging |
| url | http://www.sciencedirect.com/science/article/pii/S2213597921000689 |
| work_keys_str_mv | AT leikang deeplearningenablesultravioletphotoacousticmicroscopybasedhistologicalimagingwithnearrealtimevirtualstaining AT xiufengli deeplearningenablesultravioletphotoacousticmicroscopybasedhistologicalimagingwithnearrealtimevirtualstaining AT yanzhang deeplearningenablesultravioletphotoacousticmicroscopybasedhistologicalimagingwithnearrealtimevirtualstaining AT terencetwwong deeplearningenablesultravioletphotoacousticmicroscopybasedhistologicalimagingwithnearrealtimevirtualstaining |