A Simple Improvement to PIP-Net for Medical Image Anomaly Detection
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
The application of AI technology in domains requiring decision accountability, such as healthcare, has increased the demand for model interpretability. The part-prototype model is a well-established interpretable approach for image recognition, with PIP-Net demonstrating strong classification performance and high interpretability in multiclass classification tasks. However, PIP-Net assumes the presence of class-specific prototypes. This assumption does not hold for tasks like anomaly detection, where no local features are exclusive to the normal class. To address this, we propose an architecture that learns only the scores corresponding to the anomaly class for each prototype. This approach is based on more reasonable assumptions for anomaly detection than PIP-Net and enables concise inference using fewer prototypes. Evaluation of this approach using the MURA dataset, a large dataset of bone X-rays, revealed that the proposed architecture achieved better anomaly detection performance than the original PIP-Net with fewer prototypes.
Description
@inproceedings{10.2312:stag.20241337,
booktitle = {Smart Tools and Applications in Graphics - Eurographics Italian Chapter Conference},
editor = {Caputo, Ariel and Garro, Valeria and Giachetti, Andrea and Castellani, Umberto and Dulecha, Tinsae Gebrechristos},
title = {{A Simple Improvement to PIP-Net for Medical Image Anomaly Detection}},
author = {Kobayashi, Yuki and Yamaguchi, Yasushi},
year = {2024},
publisher = {The Eurographics Association},
ISSN = {2617-4855},
ISBN = {978-3-03868-265-3},
DOI = {10.2312/stag.20241337}
}