Diffusion Models with Implicit Guidance for Medical Anomaly Detection
This work addresses improved specificity and reduced false positives in radiological evaluations for medical imaging, representing an incremental advancement.
The paper tackles the problem of preserving healthy tissue information during pathology removal in medical anomaly detection using diffusion models, resulting in THOR, which outperforms existing methods in detecting and segmenting anomalies in brain MRIs and wrist X-rays.
Diffusion models have advanced unsupervised anomaly detection by improving the transformation of pathological images into pseudo-healthy equivalents. Nonetheless, standard approaches may compromise critical information during pathology removal, leading to restorations that do not align with unaffected regions in the original scans. Such discrepancies can inadvertently increase false positive rates and reduce specificity, complicating radiological evaluations. This paper introduces Temporal Harmonization for Optimal Restoration (THOR), which refines the de-noising process by integrating implicit guidance through temporal anomaly maps. THOR aims to preserve the integrity of healthy tissue in areas unaffected by pathology. Comparative evaluations show that THOR surpasses existing diffusion-based methods in detecting and segmenting anomalies in brain MRIs and wrist X-rays. Code: https://github.com/ci-ber/THOR_DDPM.