StyleGAN2-based Out-of-Distribution Detection for Medical Imaging

McKell Woodland, John Wood, Caleb O'Connor, Ankit B. Patel, Kristy K. Brock
7/10/2023
eess.IVcs.LG

No official code found. Best implementation: mckellwoodland/dimen_reduce_mahal (4 stars, MIT, last commit 2025-05-20).

Abstract

One barrier to the clinical deployment of deep learning-based models is the presence of images at runtime that lie far outside the training distribution of a given model. We aim to detect these out-of-distribution (OOD) images with a generative adversarial network (GAN). Our training dataset was comprised of 3,234 liver-containing computed tomography (CT) scans from 456 patients. Our OOD test data consisted of CT images of the brain, head and neck, lung, cervix, and abnormal livers. A StyleGAN2-ADA architecture was employed to model the training distribution. Images were reconstructed using backpropagation. Reconstructions were evaluated using the Wasserstein distance, mean squared error, and the structural similarity index measure. OOD detection was evaluated with the area under the receiver operating characteristic curve (AUROC). Our paradigm distinguished between liver and non-liver CT with greater than 90% AUROC. It was also completely unable to reconstruct liver artifacts, such as needles and ascites.

View on arXivView PDF

Code Implementations(1)

mckellwoodland/dimen_reduce_mahalCommunity100%
MIT

Official Repository for the MELBA paper entitled "Dimensionality Reduction and Nearest Neighbors for Improving Out-of-Distribution Detection in Medical Image Segmentation".

41Python, DockerfileAug 4, 20231 years agoMIT
k-nearest-neighborliver-cancermahalanobis-distancemedical-imaging-segmentationout-of-distribution-detection

Similar Papers

Toward a Realistic Benchmark for Out-of-Distribution Detection

84% match2024

Out-of-distribution Detection in Medical Image Analysis: A survey

83% match2024

Comparing Adversarial and Supervised Learning for Organs at Risk Segmentation in CT images

82% match2023

Are We Ready for Out-of-Distribution Detection in Digital Pathology?

82% match2024

Multi-layer Aggregation as a key to feature-based OOD detection

81% match2023

NERO: Explainable Out-of-Distribution Detection with Neuron-level Relevance

81% match1 repo2025

Cite this paper

@article{woodland2023styleganbased,
  title  = {StyleGAN2-based Out-of-Distribution Detection for Medical Imaging},
  author = {McKell Woodland and John Wood and Caleb O'Connor and Ankit B. Patel and Kristy K. Brock},
  year   = {2023},
  eprint = {2307.10193},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.10193v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.10193.svg)](https://codeofpaper.com/paper/2307.10193)

Discussion