Hybrid Perception Loss-Driven Synthetic Images Generation of Pathological Myopia Stages
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Standard
2025 IEEE 34th International Symposium on Industrial Electronics, ISIE 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (IEEE International Symposium on Industrial Electronics).
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Hybrid Perception Loss-Driven Synthetic Images Generation of Pathological Myopia Stages
AU - Herrera-Chavez, Andre I.
AU - Flores-Fuentes, Wendy
AU - Rodriguez-Quinonez, Julio C.
AU - Rodriguez-Martinez, Eder A.
AU - Sergiyenko, Oleg
AU - Mercorelli, Paolo
AU - Castro-Toscano, Moises J.
N1 - Publisher Copyright: © 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Pathological Myopia (PM) progresses through distinct stages - Tessellated Fundus, Choroidal Atrophy, and Patchy Atrophy. The limited availability of annotated datasets poses challenges for developing machine learning models tailored to these stages. This study introduces a novel framework for synthetic image generation using CycleGAN, enhanced with a hybrid perceptual loss. By leveraging a CNN-based feature extractor, this approach refines biomarker representation and ensures their preservation across PM stages while improving image quality. The hybrid perceptual loss aligns generated images with high-level features from real images, enhancing biomarker accuracy and structural fidelity. This methodology not only augments dataset diversity but also facilitates clinical applications by producing synthetic images that faithfully represent PM stages and biomarkers, contributing to the advancement of ophthalmological diagnostics.
AB - Pathological Myopia (PM) progresses through distinct stages - Tessellated Fundus, Choroidal Atrophy, and Patchy Atrophy. The limited availability of annotated datasets poses challenges for developing machine learning models tailored to these stages. This study introduces a novel framework for synthetic image generation using CycleGAN, enhanced with a hybrid perceptual loss. By leveraging a CNN-based feature extractor, this approach refines biomarker representation and ensures their preservation across PM stages while improving image quality. The hybrid perceptual loss aligns generated images with high-level features from real images, enhancing biomarker accuracy and structural fidelity. This methodology not only augments dataset diversity but also facilitates clinical applications by producing synthetic images that faithfully represent PM stages and biomarkers, contributing to the advancement of ophthalmological diagnostics.
KW - Biomarker Generation
KW - CycleGAN
KW - Machine Learning in Ophthalmology
KW - Medical Image Processing
KW - Pathological Myopia
KW - Perceptual Loss
KW - Synthetic Image Generation
KW - Engineering
UR - http://www.scopus.com/inward/record.url?scp=105016214372&partnerID=8YFLogxK
U2 - 10.1109/ISIE62713.2025.11124683
DO - 10.1109/ISIE62713.2025.11124683
M3 - Article in conference proceedings
AN - SCOPUS:105016214372
SN - 979-8-3503-7480-3
T3 - IEEE International Symposium on Industrial Electronics
BT - 2025 IEEE 34th International Symposium on Industrial Electronics, ISIE 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 34th IEEE International Symposium on Industrial Electronics, ISIE 2025
Y2 - 20 June 2025 through 23 June 2025
ER -