Benchmarking Pretrained Vision Embeddings for Near- and Duplicate Detection in Medical Images
This addresses data quality issues in medical imaging datasets, but it is incremental as it applies existing methods to a new domain.
The paper tackled the problem of near- and duplicate detection in medical images to prevent data leakage and biases, achieving a mean sensitivity of 0.9645 and specificity of 0.8559 using pretrained vision embeddings.
Near- and duplicate image detection is a critical concern in the field of medical imaging. Medical datasets often contain similar or duplicate images from various sources, which can lead to significant performance issues and evaluation biases, especially in machine learning tasks due to data leakage between training and testing subsets. In this paper, we present an approach for identifying near- and duplicate 3D medical images leveraging publicly available 2D computer vision embeddings. We assessed our approach by comparing embeddings extracted from two state-of-the-art self-supervised pretrained models and two different vector index structures for similarity retrieval. We generate an experimental benchmark based on the publicly available Medical Segmentation Decathlon dataset. The proposed method yields promising results for near- and duplicate image detection achieving a mean sensitivity and specificity of 0.9645 and 0.8559, respectively.