Multi-style conversion for semantic segmentation of lesions in fundus images by adversarial attacks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Playout, Clément, Duval, Renaud, Boucher, Marie Carole, Cheriet, Farida
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910654665850880
author Playout, Clément
Duval, Renaud
Boucher, Marie Carole
Cheriet, Farida
author_facet Playout, Clément
Duval, Renaud
Boucher, Marie Carole
Cheriet, Farida
contents The diagnosis of diabetic retinopathy, which relies on fundus images, faces challenges in achieving transparency and interpretability when using a global classification approach. However, segmentation-based databases are significantly more expensive to acquire and combining them is often problematic. This paper introduces a novel method, termed adversarial style conversion, to address the lack of standardization in annotation styles across diverse databases. By training a single architecture on combined databases, the model spontaneously modifies its segmentation style depending on the input, demonstrating the ability to convert among different labeling styles. The proposed methodology adds a linear probe to detect dataset origin based on encoder features and employs adversarial attacks to condition the model's segmentation style. Results indicate significant qualitative and quantitative through dataset combination, offering avenues for improved model generalization, uncertainty estimation and continuous interpolation between annotation styles. Our approach enables training a segmentation model with diverse databases while controlling and leveraging annotation styles for improved retinopathy diagnosis.
format Preprint
id arxiv_https___arxiv_org_abs_2410_13822
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-style conversion for semantic segmentation of lesions in fundus images by adversarial attacks
Playout, Clément
Duval, Renaud
Boucher, Marie Carole
Cheriet, Farida
Computer Vision and Pattern Recognition
Artificial Intelligence
The diagnosis of diabetic retinopathy, which relies on fundus images, faces challenges in achieving transparency and interpretability when using a global classification approach. However, segmentation-based databases are significantly more expensive to acquire and combining them is often problematic. This paper introduces a novel method, termed adversarial style conversion, to address the lack of standardization in annotation styles across diverse databases. By training a single architecture on combined databases, the model spontaneously modifies its segmentation style depending on the input, demonstrating the ability to convert among different labeling styles. The proposed methodology adds a linear probe to detect dataset origin based on encoder features and employs adversarial attacks to condition the model's segmentation style. Results indicate significant qualitative and quantitative through dataset combination, offering avenues for improved model generalization, uncertainty estimation and continuous interpolation between annotation styles. Our approach enables training a segmentation model with diverse databases while controlling and leveraging annotation styles for improved retinopathy diagnosis.
title Multi-style conversion for semantic segmentation of lesions in fundus images by adversarial attacks
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2410.13822