Ultrawidefield-to-Conventional Fundus Image Translation with Scaled Feature Registration and Distorted Vessel Correction
Bioengineering · 28.09.2025 · 12 (10) · 1046
DOI: 10.3390/bioengineering12101046 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioengineering, online erschienen am 28.09.2025. Verfasst von JuChan Kim und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of AI Systems Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Development of Brain Disease (Stroke) Prediction Model based on Fundus Image Analysis Program, the ICT Creative Consilience Program, the AI Innovation Hub.
Abstract (Originalsprache)
Conventional fundus (CF) and ultrawidefield fundus (UF) imaging are two primary modalities widely used in ophthalmology. Despite the complementary use of both imaging modalities in clinical practice, existing research on fundus image translation has yet to reach clinical viability and often lacks the necessary accuracy and detail required for practical medical use. Additionally, collecting paired UFI-CFI data from the same patients presents significant limitations, and unpaired learning-based generative models frequently suffer from distortion phenomena, such as hallucinations. This study introduces an enhanced modality transformation method to improve the diagnostic support capabilities of deep learning models in ophthalmology. The proposed method translates UF images (UFIs) into CF images (CFIs), potentially replacing the dual-imaging approach commonly used in clinical practice. This replacement can significantly reduce financial and temporal burdens on patients. To achieve this, this study leveraged UFI–CFI image pairs obtained from the same patient on the same day. This approach minimizes information distortion and accurately converts the two modalities. Our model employs scaled feature registration and distorted vessel correction methods to align UFI–CFI pairs effectively. The generated CFIs not only enhance image quality and better represent the retinal area compared to existing methods but also effectively preserve disease-related details from UFIs, aiding in accurate diagnosis. Furthermore, compared with existing methods, our model demonstrated a substantial 18.2% reduction in MSE, a 7.2% increase in PSNR, and a 12.7% improvement in SSIM metrics. Notably, our results show that the generated CFIs are nearly indistinguishable from the real CFIs, as confirmed by ophthalmologists.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of AI Systems Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea.
- Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon 16419, Republic of Korea.
- Department of Software Science and Engineering, Kunsan National University, Gunsan 54150, Republic of Korea.
- Department of Ophthalmology, Samsung Medical Center, Seoul 06351, Republic of Korea.
- School of Medicine, Sungkyunkwan University, Suwon 16419, Republic of Korea.
- Samsung Advanced Institute for Health Sciences and Technology, Sungkyunkwan University, Suwon 16419, Republic of Korea.
ORCID-Profile: JuChan Kim, Junghyun Bum, Hyunseung Choo
Förderung
- the Development of Brain Disease (Stroke) Prediction Model based on Fundus Image Analysis Program
- the ICT Creative Consilience Program
- the AI Innovation Hub
- the Korea government
Zitieren
APA
Kim, J., Bum, J., Le, D. T., Son, C. H., Lee, E. J., Han, J. C., & Choo, H. (2025). Ultrawidefield-to-Conventional Fundus Image Translation with Scaled Feature Registration and Distorted Vessel Correction. Bioengineering, 12(10), 1046. https://doi.org/10.3390/bioengineering12101046
Vancouver
Kim J, Bum J, Le DT, Son CH, Lee EJ, Han JC, et al. Ultrawidefield-to-Conventional Fundus Image Translation with Scaled Feature Registration and Distorted Vessel Correction. Bioengineering. 2025;12(10):1046. doi:10.3390/bioengineering12101046
BibTeX
@article{Kim2025,
author = {Kim, JuChan and Bum, Junghyun and Le, Duc-Tai and Son, Chang-Hwan and Lee, Eun Jung and Han, Jong Chul and Choo, Hyunseung},
title = {Ultrawidefield-to-Conventional Fundus Image Translation with Scaled Feature Registration and Distorted Vessel Correction},
journal = {Bioengineering},
year = {2025},
volume = {12},
number = {10},
pages = {1046},
doi = {10.3390/bioengineering12101046}
}
EUVIMED