MIAS-SAM: Medical Image Anomaly Segmentation without thresholding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Colussi, Marco, Ahmetovic, Dragan, Mascetti, Sergio
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915311444295680
author Colussi, Marco
Ahmetovic, Dragan
Mascetti, Sergio
author_facet Colussi, Marco
Ahmetovic, Dragan
Mascetti, Sergio
contents This paper presents MIAS-SAM, a novel approach for the segmentation of anomalous regions in medical images. MIAS-SAM uses a patch-based memory bank to store relevant image features, which are extracted from normal data using the SAM encoder. At inference time, the embedding patches extracted from the SAM encoder are compared with those in the memory bank to obtain the anomaly map. Finally, MIAS-SAM computes the center of gravity of the anomaly map to prompt the SAM decoder, obtaining an accurate segmentation from the previously extracted features. Differently from prior works, MIAS-SAM does not require to define a threshold value to obtain the segmentation from the anomaly map. Experimental results conducted on three publicly available datasets, each with a different imaging modality (Brain MRI, Liver CT, and Retina OCT) show accurate anomaly segmentation capabilities measured using DICE score. The code is available at: https://github.com/warpcut/MIAS-SAM
format Preprint
id arxiv_https___arxiv_org_abs_2505_22762
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MIAS-SAM: Medical Image Anomaly Segmentation without thresholding
Colussi, Marco
Ahmetovic, Dragan
Mascetti, Sergio
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
This paper presents MIAS-SAM, a novel approach for the segmentation of anomalous regions in medical images. MIAS-SAM uses a patch-based memory bank to store relevant image features, which are extracted from normal data using the SAM encoder. At inference time, the embedding patches extracted from the SAM encoder are compared with those in the memory bank to obtain the anomaly map. Finally, MIAS-SAM computes the center of gravity of the anomaly map to prompt the SAM decoder, obtaining an accurate segmentation from the previously extracted features. Differently from prior works, MIAS-SAM does not require to define a threshold value to obtain the segmentation from the anomaly map. Experimental results conducted on three publicly available datasets, each with a different imaging modality (Brain MRI, Liver CT, and Retina OCT) show accurate anomaly segmentation capabilities measured using DICE score. The code is available at: https://github.com/warpcut/MIAS-SAM
title MIAS-SAM: Medical Image Anomaly Segmentation without thresholding
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2505.22762