EUVIMEDEUVIMED entdecken
Crossref

IQE-CLIP: Instance-Aware Query Embedding for Zero-/Few-Shot Anomaly Detection in Medical Domain

Hong Huang, Weixiang Sun, Zhijian Wu, Donghuan Lu, Xian Wu, Yefeng Zheng

IEEE Transactions on Image Processing · 2026 · 35 · 6153-6166

DOI: 10.1109/tip.2026.3697648 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2026. Verfasst von Hong Huang und 5 weiteren Autor:innen. Die Arbeit nennt 73 Literaturangaben. Gefördert von: Zhejiang Leading Innovative and Entrepreneur Team Introduction Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recently, the rapid advancements of vision-language models, such as CLIP, have led to significant progress in zero-/few-shot anomaly detection (ZFSAD) tasks. However, most existing CLIP-based ZFSAD methods commonly assume prior knowledge of categories and rely on carefully crafted prompts tailored to specific scenarios. While such meticulously designed text prompts effectively capture semantic information in the textual space, they fall short of distinguishing normal and anomalous instances within the joint embedding space. Moreover, these ZFSAD methods are predominantly explored in industrial scenarios, with few efforts conducted for medical tasks. To this end, we propose an innovative framework for ZFSAD tasks in the medical domain, denoted as IQE-CLIP. We reveal that query embeddings, which incorporate both textual and instance-aware visual information, are better indicators for abnormalities. Specifically, we first introduce class-based prompting tokens and learnable prompting tokens for better adaptation of CLIP to the medical domain. Then, we design an instance-aware query module (IQM) to extract region-level contextual information from both text prompts and visual features, enabling the generation of query embeddings that are more sensitive to anomalies. Extensive experiments conducted on six medical datasets demonstrate that IQE-CLIP achieves state-of-the-art performance on both zero-shot and few-shot tasks. The source code and data are available at https://github.com/hongh0/IQE-CLIP.

Förderung

  • Zhejiang Leading Innovative and Entrepreneur Team Introduction Program

Zitieren

APA

Huang, H., Sun, W., Wu, Z., Lu, D., Wu, X., & Zheng, Y. (2026). IQE-CLIP: Instance-Aware Query Embedding for Zero-/Few-Shot Anomaly Detection in Medical Domain. IEEE Transactions on Image Processing, 35, 6153-6166. https://doi.org/10.1109/tip.2026.3697648

Vancouver

Huang H, Sun W, Wu Z, Lu D, Wu X, Zheng Y. IQE-CLIP: Instance-Aware Query Embedding for Zero-/Few-Shot Anomaly Detection in Medical Domain. IEEE Transactions on Image Processing. 2026;35:6153-6166. doi:10.1109/tip.2026.3697648

BibTeX
@article{Huang2026,
  author = {Huang, Hong and Sun, Weixiang and Wu, Zhijian and Lu, Donghuan and Wu, Xian and Zheng, Yefeng},
  title = {IQE-CLIP: Instance-Aware Query Embedding for Zero-/Few-Shot Anomaly Detection in Medical Domain},
  journal = {IEEE Transactions on Image Processing},
  year = {2026},
  volume = {35},
  pages = {6153--6166},
  doi = {10.1109/tip.2026.3697648}
}
Kostenlos starten →