Toward Foundation Detection Models via Example-Based Grounding in EM-DETR
arXiv cs.AI · 11 Nov 2026, 04:59 AEST
A novel approach to building a multi-domain foundation model for lesion detection has been introduced. This model integrates diverse medical image datasets and achieves state-of-the-art results in medical imaging.
Key facts
- DeepLearning (DL)-based Computer-Aided Diagnostic (CAD) systems have shown great promise in supporting clinicians, but their performance heavily depends on the availability of large, well-annotated datasets.
- In this work, we introduce a novel approach to building a multi-domain foundation model for lesion detection by integrating diverse medical image datasets annotated for both anatomical structures and abnormalities.
- Our model achieves state-of-the-art (SOTA) results, with image-level AUC scores exceeding previous SOTA by 1% point on four key findings in CXRs.
- We achieve a leading mAP50 in CXRs and mammography, surpassing EM-DETR trained on a single dataset by 2-3% points.
Read aloud
A novel approach to building a multi-domain foundation model for lesion detection has been introduced. This model integrates diverse medical image datasets and achieves state-of-the-art results in medical imaging. DeepLearning DL -based Computer-Aided Diagnostic CAD systems have shown great promise in supporting clinicians, but their performance heavily depends on the availability of large, well-annotated datasets. In this work, we introduce a novel approach to building a multi-domain foundation model for lesion detection by integrating diverse medical image datasets annotated for both anatomical structures and abnormalities. Our model achieves state-of-the-art SOTA results, with image-level AUC scores exceeding previous SOTA by one percent point on four key findings in CXRs. We achieve a leading mAP50 in CXRs and mammography, surpassing EM-DETR trained on a single dataset by two-three percent points.
Sources in this cluster
- Toward Foundation Detection Models via Example-Based Grounding in EM-DETR
arXiv cs.AI · 11 Nov 2026, 04:59 AEST