Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis
Abstract Background Due to the complexity and cost of preparing histopathological slides, deep learning-based methods have been developed to generate high-quality histological images. However, existing approaches primarily focus on spatial domain information, neglecting the periodic information in t...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-025-06057-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571226644545536 |
|---|---|
| author | Qifeng Liu Tao Zhou Chi Cheng Jin Ma Marzia Hoque Tania |
| author_facet | Qifeng Liu Tao Zhou Chi Cheng Jin Ma Marzia Hoque Tania |
| author_sort | Qifeng Liu |
| collection | DOAJ |
| description | Abstract Background Due to the complexity and cost of preparing histopathological slides, deep learning-based methods have been developed to generate high-quality histological images. However, existing approaches primarily focus on spatial domain information, neglecting the periodic information in the frequency domain and the complementary relationship between the two domains. In this paper, we proposed a generative adversarial network that employs a cross-attention mechanism to extract and fuse features across spatial and frequency domains. The method optimizes frequency domain features using spatial domain guidance and refines spatial features with frequency domain information, preserving key details while eliminating redundancy to generate high-quality histological images. Results Our model incorporates a variable-window mixed attention module to dynamically adjust attention window sizes, capturing both local details and global context. A spectral filtering module enhances the extraction of repetitive textures and periodic structures, while a cross-attention fusion module dynamically weights features from both domains, focusing on the most critical information to produce realistic and detailed images. Conclusions The proposed method achieves efficient spatial-frequency domain fusion, significantly improving image generation quality. Experiments on the Patch Camelyon dataset show superior performance over eight state-of-the-art models across five metrics. This approach advances automated histopathological image generation with potential for clinical applications. |
| format | Article |
| id | doaj-art-3c7bf3f84f6140218158c032adc0535e |
| institution | Kabale University |
| issn | 1471-2105 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj-art-3c7bf3f84f6140218158c032adc0535e2025-02-02T12:45:01ZengBMCBMC Bioinformatics1471-21052025-01-0126112310.1186/s12859-025-06057-9Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesisQifeng Liu0Tao Zhou1Chi Cheng2Jin Ma3Marzia Hoque Tania4Centre for Big Data Research in Health, University of New South WalesDepartment of Respiratory and Critical Medicine, The Second Affiliated Hospital of Nanchang UniversitySchool of Computer Science and Engineering, University of New South WalesFaculty of Engineering, The University of SydneyCentre for Big Data Research in Health, University of New South WalesAbstract Background Due to the complexity and cost of preparing histopathological slides, deep learning-based methods have been developed to generate high-quality histological images. However, existing approaches primarily focus on spatial domain information, neglecting the periodic information in the frequency domain and the complementary relationship between the two domains. In this paper, we proposed a generative adversarial network that employs a cross-attention mechanism to extract and fuse features across spatial and frequency domains. The method optimizes frequency domain features using spatial domain guidance and refines spatial features with frequency domain information, preserving key details while eliminating redundancy to generate high-quality histological images. Results Our model incorporates a variable-window mixed attention module to dynamically adjust attention window sizes, capturing both local details and global context. A spectral filtering module enhances the extraction of repetitive textures and periodic structures, while a cross-attention fusion module dynamically weights features from both domains, focusing on the most critical information to produce realistic and detailed images. Conclusions The proposed method achieves efficient spatial-frequency domain fusion, significantly improving image generation quality. Experiments on the Patch Camelyon dataset show superior performance over eight state-of-the-art models across five metrics. This approach advances automated histopathological image generation with potential for clinical applications.https://doi.org/10.1186/s12859-025-06057-9Generative adversarial networksCross-attention mechanismSpatial domainFrequency domainHistological slide imagesVariable-window mixed attention |
| spellingShingle | Qifeng Liu Tao Zhou Chi Cheng Jin Ma Marzia Hoque Tania Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis BMC Bioinformatics Generative adversarial networks Cross-attention mechanism Spatial domain Frequency domain Histological slide images Variable-window mixed attention |
| title | Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| title_full | Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| title_fullStr | Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| title_full_unstemmed | Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| title_short | Hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| title_sort | hybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis |
| topic | Generative adversarial networks Cross-attention mechanism Spatial domain Frequency domain Histological slide images Variable-window mixed attention |
| url | https://doi.org/10.1186/s12859-025-06057-9 |
| work_keys_str_mv | AT qifengliu hybridgenerativeadversarialnetworkbasedonfrequencyandspatialdomainforhistopathologicalimagesynthesis AT taozhou hybridgenerativeadversarialnetworkbasedonfrequencyandspatialdomainforhistopathologicalimagesynthesis AT chicheng hybridgenerativeadversarialnetworkbasedonfrequencyandspatialdomainforhistopathologicalimagesynthesis AT jinma hybridgenerativeadversarialnetworkbasedonfrequencyandspatialdomainforhistopathologicalimagesynthesis AT marziahoquetania hybridgenerativeadversarialnetworkbasedonfrequencyandspatialdomainforhistopathologicalimagesynthesis |