The reviewed record of science sign in
Pith

arxiv: 2310.03664 · v1 · pith:4FB6WEI2 · submitted 2023-10-05 · eess.IV · cs.CV

Certification of Deep Learning Models for Medical Image Segmentation

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:4FB6WEI2record.jsonopen to challenge →

classification eess.IV cs.CV
keywords modelssegmentationmedicalrandomizedsmoothingcertifiedcertifydiffusion
0
0 comments X
read the original abstract

In medical imaging, segmentation models have known a significant improvement in the past decade and are now used daily in clinical practice. However, similar to classification models, segmentation models are affected by adversarial attacks. In a safety-critical field like healthcare, certifying model predictions is of the utmost importance. Randomized smoothing has been introduced lately and provides a framework to certify models and obtain theoretical guarantees. In this paper, we present for the first time a certified segmentation baseline for medical imaging based on randomized smoothing and diffusion models. Our results show that leveraging the power of denoising diffusion probabilistic models helps us overcome the limits of randomized smoothing. We conduct extensive experiments on five public datasets of chest X-rays, skin lesions, and colonoscopies, and empirically show that we are able to maintain high certified Dice scores even for highly perturbed images. Our work represents the first attempt to certify medical image segmentation models, and we aspire for it to set a foundation for future benchmarks in this crucial and largely uncharted area.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.