Autofluorescence Imaging and Virtual Histological Staining of Human Prostate Sections for Cancer Diagnosis

Mingxuan SI; Weixing DAI; Ivy H.M. Wong; Yan ZHANG; Tony K.F. Ma; Ping Wing Ng; Kwok Hung Li; Sylvia M.S. Wong; John H.K. Ngan; Terence T.W. Wong

doi:10.1002/aisy.202401081

Autofluorescence Imaging and Virtual Histological Staining of Human Prostate Sections for Cancer Diagnosis

Mingxuan SI, Weixing DAI, Ivy H.M. Wong, Yan ZHANG, Tony K.F. Ma, Ping Wing Ng, Kwok Hung Li, Sylvia M.S. Wong, John H.K. Ngan^*, Terence T.W. Wong^*

^*Corresponding author for this work

Department of Chemical and Biological Engineering

Research output: Contribution to journal › Journal Article › peer-review

Abstract

Autofluorescence imaging with virtual histological staining has shown promising potential to improve the efficiency of postoperative histopathology workflow. Here, a widefield deep-ultraviolet light-emitting diode-based imaging system to generate autofluorescence images of human prostate tissue sections from 12 patients with a lateral resolution of ≈1 μm is used. Subsequently, the autofluorescence images are transformed into virtual hematoxylin and eosin-stained images via a weakly supervised deep learning framework. The virtual staining results are assessed by four professional pathologists through statistical analysis, which shows high diagnostic accuracy (91%), and high tissue detail quality.

Original language	English
Article number	2401081
Journal	Advanced Intelligent Systems
DOIs	https://doi.org/10.1002/aisy.202401081
Publication status	Accepted/In press - 20 May 2025

Bibliographical note

Publisher Copyright:
© 2025 The Author(s). Advanced Intelligent Systems published by Wiley-VCH GmbH.

Keywords

autofluorescences
deep learning
histopathologies
prostate cancers
ultraviolet

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1002/aisy.202401081Licence: CC BY

Cite this

@article{864c8e8a4c4a4fe78af75b60bdd1ed16,

title = "Autofluorescence Imaging and Virtual Histological Staining of Human Prostate Sections for Cancer Diagnosis",

abstract = "Autofluorescence imaging with virtual histological staining has shown promising potential to improve the efficiency of postoperative histopathology workflow. Here, a widefield deep-ultraviolet light-emitting diode-based imaging system to generate autofluorescence images of human prostate tissue sections from 12 patients with a lateral resolution of ≈1 μm is used. Subsequently, the autofluorescence images are transformed into virtual hematoxylin and eosin-stained images via a weakly supervised deep learning framework. The virtual staining results are assessed by four professional pathologists through statistical analysis, which shows high diagnostic accuracy (91\%), and high tissue detail quality.",

keywords = "autofluorescences, deep learning, histopathologies, prostate cancers, ultraviolet",

author = "Mingxuan SI and Weixing DAI and Wong, \{Ivy H.M.\} and Yan ZHANG and Ma, \{Tony K.F.\} and Ng, \{Ping Wing\} and Li, \{Kwok Hung\} and Wong, \{Sylvia M.S.\} and Ngan, \{John H.K.\} and Wong, \{Terence T.W.\}",

note = "Publisher Copyright: {\textcopyright} 2025 The Author(s). Advanced Intelligent Systems published by Wiley-VCH GmbH.",

year = "2025",

month = may,

day = "20",

doi = "10.1002/aisy.202401081",

language = "English",

journal = "Advanced Intelligent Systems",

issn = "2640-4567",

publisher = "John Wiley and Sons Inc",

}

TY - JOUR

T1 - Autofluorescence Imaging and Virtual Histological Staining of Human Prostate Sections for Cancer Diagnosis

AU - SI, Mingxuan

AU - DAI, Weixing

AU - Wong, Ivy H.M.

AU - ZHANG, Yan

AU - Ma, Tony K.F.

AU - Ng, Ping Wing

AU - Li, Kwok Hung

AU - Wong, Sylvia M.S.

AU - Ngan, John H.K.

AU - Wong, Terence T.W.

PY - 2025/5/20

Y1 - 2025/5/20

N2 - Autofluorescence imaging with virtual histological staining has shown promising potential to improve the efficiency of postoperative histopathology workflow. Here, a widefield deep-ultraviolet light-emitting diode-based imaging system to generate autofluorescence images of human prostate tissue sections from 12 patients with a lateral resolution of ≈1 μm is used. Subsequently, the autofluorescence images are transformed into virtual hematoxylin and eosin-stained images via a weakly supervised deep learning framework. The virtual staining results are assessed by four professional pathologists through statistical analysis, which shows high diagnostic accuracy (91%), and high tissue detail quality.

AB - Autofluorescence imaging with virtual histological staining has shown promising potential to improve the efficiency of postoperative histopathology workflow. Here, a widefield deep-ultraviolet light-emitting diode-based imaging system to generate autofluorescence images of human prostate tissue sections from 12 patients with a lateral resolution of ≈1 μm is used. Subsequently, the autofluorescence images are transformed into virtual hematoxylin and eosin-stained images via a weakly supervised deep learning framework. The virtual staining results are assessed by four professional pathologists through statistical analysis, which shows high diagnostic accuracy (91%), and high tissue detail quality.

KW - autofluorescences

KW - deep learning

KW - histopathologies

KW - prostate cancers

KW - ultraviolet

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:001491469500001

UR - https://openalex.org/W4410535219

UR - https://www.scopus.com/pages/publications/105005553576

U2 - 10.1002/aisy.202401081

DO - 10.1002/aisy.202401081

M3 - Journal Article

SN - 2640-4567

JO - Advanced Intelligent Systems

JF - Advanced Intelligent Systems

M1 - 2401081

ER -

Autofluorescence Imaging and Virtual Histological Staining of Human Prostate Sections for Cancer Diagnosis

Abstract

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this