Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning
Abstract Mid-infrared photoacoustic microscopy can capture biochemical information without staining. However, the long mid-infrared optical wavelengths make the spatial resolution of photoacoustic microscopy significantly poorer than that of conventional confocal fluorescence microscopy. Here, we de...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-55262-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559272978120704 |
|---|---|
| author | Eunwoo Park Sampa Misra Dong Gyu Hwang Chiho Yoon Joongho Ahn Donggyu Kim Jinah Jang Chulhong Kim |
| author_facet | Eunwoo Park Sampa Misra Dong Gyu Hwang Chiho Yoon Joongho Ahn Donggyu Kim Jinah Jang Chulhong Kim |
| author_sort | Eunwoo Park |
| collection | DOAJ |
| description | Abstract Mid-infrared photoacoustic microscopy can capture biochemical information without staining. However, the long mid-infrared optical wavelengths make the spatial resolution of photoacoustic microscopy significantly poorer than that of conventional confocal fluorescence microscopy. Here, we demonstrate an explainable deep learning-based unsupervised inter-domain transformation of low-resolution unlabeled mid-infrared photoacoustic microscopy images into confocal-like virtually fluorescence-stained high-resolution images. The explainable deep learning-based framework is proposed for this transformation, wherein an unsupervised generative adversarial network is primarily employed and then a saliency constraint is added for better explainability. We validate the performance of explainable deep learning-based mid-infrared photoacoustic microscopy by identifying cell nuclei and filamentous actins in cultured human cardiac fibroblasts and matching them with the corresponding CFM images. The XDL ensures similar saliency between the two domains, making the transformation process more stable and more reliable than existing networks. Our XDL-MIR-PAM enables label-free high-resolution duplexed cellular imaging, which can significantly benefit many research avenues in cell biology. |
| format | Article |
| id | doaj-art-34e041ae4bde4345a1b8e644e7a53482 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-34e041ae4bde4345a1b8e644e7a534822025-01-05T12:36:49ZengNature PortfolioNature Communications2041-17232024-12-0115111210.1038/s41467-024-55262-2Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learningEunwoo Park0Sampa Misra1Dong Gyu Hwang2Chiho Yoon3Joongho Ahn4Donggyu Kim5Jinah Jang6Chulhong Kim7Department of Convergence IT Engineering, Pohang University of Science and Technology (POSTECH)Department of Convergence IT Engineering, Pohang University of Science and Technology (POSTECH)Medical Device Innovation Center, Pohang University of Science and Technology (POSTECH)Medical Device Innovation Center, Pohang University of Science and Technology (POSTECH)Medical Device Innovation Center, Pohang University of Science and Technology (POSTECH)Department of Convergence IT Engineering, Pohang University of Science and Technology (POSTECH)Department of Convergence IT Engineering, Pohang University of Science and Technology (POSTECH)Department of Convergence IT Engineering, Pohang University of Science and Technology (POSTECH)Abstract Mid-infrared photoacoustic microscopy can capture biochemical information without staining. However, the long mid-infrared optical wavelengths make the spatial resolution of photoacoustic microscopy significantly poorer than that of conventional confocal fluorescence microscopy. Here, we demonstrate an explainable deep learning-based unsupervised inter-domain transformation of low-resolution unlabeled mid-infrared photoacoustic microscopy images into confocal-like virtually fluorescence-stained high-resolution images. The explainable deep learning-based framework is proposed for this transformation, wherein an unsupervised generative adversarial network is primarily employed and then a saliency constraint is added for better explainability. We validate the performance of explainable deep learning-based mid-infrared photoacoustic microscopy by identifying cell nuclei and filamentous actins in cultured human cardiac fibroblasts and matching them with the corresponding CFM images. The XDL ensures similar saliency between the two domains, making the transformation process more stable and more reliable than existing networks. Our XDL-MIR-PAM enables label-free high-resolution duplexed cellular imaging, which can significantly benefit many research avenues in cell biology.https://doi.org/10.1038/s41467-024-55262-2 |
| spellingShingle | Eunwoo Park Sampa Misra Dong Gyu Hwang Chiho Yoon Joongho Ahn Donggyu Kim Jinah Jang Chulhong Kim Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning Nature Communications |
| title | Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning |
| title_full | Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning |
| title_fullStr | Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning |
| title_full_unstemmed | Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning |
| title_short | Unsupervised inter-domain transformation for virtually stained high-resolution mid-infrared photoacoustic microscopy using explainable deep learning |
| title_sort | unsupervised inter domain transformation for virtually stained high resolution mid infrared photoacoustic microscopy using explainable deep learning |
| url | https://doi.org/10.1038/s41467-024-55262-2 |
| work_keys_str_mv | AT eunwoopark unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT sampamisra unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT donggyuhwang unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT chihoyoon unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT joonghoahn unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT donggyukim unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT jinahjang unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning AT chulhongkim unsupervisedinterdomaintransformationforvirtuallystainedhighresolutionmidinfraredphotoacousticmicroscopyusingexplainabledeeplearning |