Contrastive Language Prompting to Ease False Positives in Medical Anomaly Detection
This work addresses false positives in medical anomaly detection, which is crucial for improving diagnostic accuracy in healthcare, though it appears incremental as it builds on existing CLIP variants.
The paper tackled the problem of false positives in medical anomaly detection by introducing the Contrastive Language Prompting (CLAP) method, which uses positive and negative text prompts to enhance detection performance, as demonstrated on the BMAD dataset with six biomedical benchmarks.
A pre-trained visual-language model, contrastive language-image pre-training (CLIP), successfully accomplishes various downstream tasks with text prompts, such as finding images or localizing regions within the image. Despite CLIP's strong multi-modal data capabilities, it remains limited in specialized environments, such as medical applications. For this purpose, many CLIP variants-i.e., BioMedCLIP, and MedCLIP-SAMv2-have emerged, but false positives related to normal regions persist. Thus, we aim to present a simple yet important goal of reducing false positives in medical anomaly detection. We introduce a Contrastive LAnguage Prompting (CLAP) method that leverages both positive and negative text prompts. This straightforward approach identifies potential lesion regions by visual attention to the positive prompts in the given image. To reduce false positives, we attenuate attention on normal regions using negative prompts. Extensive experiments with the BMAD dataset, including six biomedical benchmarks, demonstrate that CLAP method enhances anomaly detection performance. Our future plans include developing an automated fine prompting method for more practical usage.