Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation
Generative Adversarial Networks (GANs) are capturing the attention of peer researchers in paired or unpaired image-to-image translation applications, particularly in the domain of retinal image processing. Additionally, there are several effective image preprocessing techniques available that can si...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Computer Methods and Programs in Biomedicine Update |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666990025000035 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576409700139008 |
|---|---|
| author | Veena K.M. Veena Mayya Rashmi Naveen Raj Sulatha V. Bhandary Uma Kulkarni |
| author_facet | Veena K.M. Veena Mayya Rashmi Naveen Raj Sulatha V. Bhandary Uma Kulkarni |
| author_sort | Veena K.M. |
| collection | DOAJ |
| description | Generative Adversarial Networks (GANs) are capturing the attention of peer researchers in paired or unpaired image-to-image translation applications, particularly in the domain of retinal image processing. Additionally, there are several effective image preprocessing techniques available that can significantly improve the performance of GANs. This study examines the impact of five different image preprocessing techniques - Green Channel, CLAHE on Green Channel, CLAHE on RGB channels, Green Channel Gaussian Convolution, and RGB Gaussian Convolution - on five different GAN variants: CycleGAN, Pix2Pix GAN, CUT GAN, FastCut GAN, and NICE GAN. The study involved conducting 30 experiments to assess the performances of these GAN variants in the image-to-image translation of dual-mode retinal images. The evaluation utilized Frechet Inception Distance (FID) and Kernel Inception Distance (KID) metric scores to measure the performance of the GAN variants. The results demonstrated that the CycleGAN model achieved the best performance with CLAHE on RGB preprocessed images, achieving the lowest FID and KID scores of 103.49 and 0.038, respectively. This investigation underscores the significant potential of image preprocessing techniques in enhancing the performance of GANs in image translation applications. |
| format | Article |
| id | doaj-art-60a9364775244c05b1c20b01ee6284b5 |
| institution | Kabale University |
| issn | 2666-9900 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Computer Methods and Programs in Biomedicine Update |
| spelling | doaj-art-60a9364775244c05b1c20b01ee6284b52025-01-31T05:12:36ZengElsevierComputer Methods and Programs in Biomedicine Update2666-99002025-01-017100179Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translationVeena K.M.0Veena Mayya1Rashmi Naveen Raj2Sulatha V. Bhandary3Uma Kulkarni4Department of Information & Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, IndiaDepartment of Information & Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India; Corresponding authors.Department of Information & Communication Technology, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India; Corresponding authors.Department of Ophthalmology, Kasturba Medical College Manipal, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, IndiaDepartment of Ophthalmology, Yenepoya Medical College, Yenepoya (Deemed to be) University, Mangalore, 575018, Karnataka, IndiaGenerative Adversarial Networks (GANs) are capturing the attention of peer researchers in paired or unpaired image-to-image translation applications, particularly in the domain of retinal image processing. Additionally, there are several effective image preprocessing techniques available that can significantly improve the performance of GANs. This study examines the impact of five different image preprocessing techniques - Green Channel, CLAHE on Green Channel, CLAHE on RGB channels, Green Channel Gaussian Convolution, and RGB Gaussian Convolution - on five different GAN variants: CycleGAN, Pix2Pix GAN, CUT GAN, FastCut GAN, and NICE GAN. The study involved conducting 30 experiments to assess the performances of these GAN variants in the image-to-image translation of dual-mode retinal images. The evaluation utilized Frechet Inception Distance (FID) and Kernel Inception Distance (KID) metric scores to measure the performance of the GAN variants. The results demonstrated that the CycleGAN model achieved the best performance with CLAHE on RGB preprocessed images, achieving the lowest FID and KID scores of 103.49 and 0.038, respectively. This investigation underscores the significant potential of image preprocessing techniques in enhancing the performance of GANs in image translation applications.http://www.sciencedirect.com/science/article/pii/S2666990025000035Medical imagingClinical decision support systemsRetinal imagingFluorescein angiographyGenerative adversarial networksImage preprocessing |
| spellingShingle | Veena K.M. Veena Mayya Rashmi Naveen Raj Sulatha V. Bhandary Uma Kulkarni Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation Computer Methods and Programs in Biomedicine Update Medical imaging Clinical decision support systems Retinal imaging Fluorescein angiography Generative adversarial networks Image preprocessing |
| title | Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation |
| title_full | Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation |
| title_fullStr | Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation |
| title_full_unstemmed | Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation |
| title_short | Analysis of preprocessing for Generative Adversarial Networks: A case study on color fundoscopy to fluorescein angiography image-to-image translation |
| title_sort | analysis of preprocessing for generative adversarial networks a case study on color fundoscopy to fluorescein angiography image to image translation |
| topic | Medical imaging Clinical decision support systems Retinal imaging Fluorescein angiography Generative adversarial networks Image preprocessing |
| url | http://www.sciencedirect.com/science/article/pii/S2666990025000035 |
| work_keys_str_mv | AT veenakm analysisofpreprocessingforgenerativeadversarialnetworksacasestudyoncolorfundoscopytofluoresceinangiographyimagetoimagetranslation AT veenamayya analysisofpreprocessingforgenerativeadversarialnetworksacasestudyoncolorfundoscopytofluoresceinangiographyimagetoimagetranslation AT rashminaveenraj analysisofpreprocessingforgenerativeadversarialnetworksacasestudyoncolorfundoscopytofluoresceinangiographyimagetoimagetranslation AT sulathavbhandary analysisofpreprocessingforgenerativeadversarialnetworksacasestudyoncolorfundoscopytofluoresceinangiographyimagetoimagetranslation AT umakulkarni analysisofpreprocessingforgenerativeadversarialnetworksacasestudyoncolorfundoscopytofluoresceinangiographyimagetoimagetranslation |