Deep learning-assisted co-registration of full-spectral autofluorescence lifetime microscopic images with H&E-stained histology images

Autofluorescence lifetime images reveal unique characteristics of endogenous fluorescence in biological samples. Comprehensive understanding and clinical diagnosis rely on co-registration with the gold standard, histology images, which is extremely challenging due to the difference of both images. H...

Full description

Saved in:
Bibliographic Details
Published in:Communications biology 2022-10, Vol.5 (1), p.1119-1119, Article 1119
Main Authors: Wang, Qiang, Fernandes, Susan, Williams, Gareth O. S., Finlayson, Neil, Akram, Ahsan R., Dhaliwal, Kevin, Hopgood, James R., Vallejo, Marta
Format: Article
Language:English
Subjects:
14/19
631/114/1305
631/114/1564
631/67/1612/1350
Biomedical and Life Sciences
Deep Learning
Fluorescence
Histology
Life Sciences
Lung cancer
Registration
Staining and Labeling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Autofluorescence lifetime images reveal unique characteristics of endogenous fluorescence in biological samples. Comprehensive understanding and clinical diagnosis rely on co-registration with the gold standard, histology images, which is extremely challenging due to the difference of both images. Here, we show an unsupervised image-to-image translation network that significantly improves the success of the co-registration using a conventional optimisation-based regression network, applicable to autofluorescence lifetime images at different emission wavelengths. A preliminary blind comparison by experienced researchers shows the superiority of our method on co-registration. The results also indicate that the approach is applicable to various image formats, like fluorescence in-tensity images. With the registration, stitching outcomes illustrate the distinct differences of the spectral lifetime across an unstained tissue, enabling macro-level rapid visual identification of lung cancer and cellular-level characterisation of cell variants and common types. The approach could be effortlessly extended to lifetime images beyond this range and other staining technologies. Using unsupervised image-to-image synthesis, homography regression-based co-registration of fluorescence-histology images is improved and can be used on various image formats across full-spectral emission wavelengths.
ISSN:2399-3642
2399-3642
DOI:10.1038/s42003-022-04090-5