Thyroid nodule segmentation improved with lightweight semantic guided AI
ThyCLIPNet: A BiomedCLIP-Guided Lightweight Attention-Enhanced DeepLabV3+ Framework for Robust Thyroid Nodule Segmentation
Computer Vision and Pattern Recognition
Summary
Detecting thyroid nodules in ultrasound images is hard because the images are blurry and noisy. The paper presents ThyCLIPNet, a computer program that uses smart image features combined with global medical knowledge to better identify nodules. This program is designed to be both accurate and efficient, working well across several thyroid image datasets. The authors show that integrating broad medical image understanding helps the program focus better on the nodules.
What this means in practice
- •For medical imaging engineers: Develop ultrasound software that segments thyroid nodules accurately and efficiently using global biomedical semantic guidance.
- •For radiology device developers: Embed lightweight and robust thyroid nodule segmentation modules into portable ultrasound devices for improved clinical diagnostics.$Commercial implications: This enables manufacturers to sell ultrasound devices with advanced nodule detection features that enhance diagnostic reliability.
Authors
Tasnim Jahan, Md Easin Arafat, Swakkhar Shatabda
Abstract
Accurate thyroid ultrasound segmentation is often challenged by low contrast, speckle noise, and unclear boundaries. Although recent methods have improved segmentation accuracy, many rely on resource-intensive architectures or lack explicit integration of multiscale features with global biomedical visual guidance. In this paper, we introduce ThyCLIPNet, a lightweight semantic-guided hybrid encoder-decoder framework that integrates BiomedCLIP-derived biomedical semantic guidance into a lightweight multi-scale CNN segmentation pipeline. The encoder integrates MobileNetV2 with efficient channel attention, while atrous spatial pyramid pooling and a custom convolutional block attention module enrich bottleneck features. The decoder combines hierarchical skip connections and lightweight attention refinement with a BiomedCLIP-guided gated fusion pathway that projects vision-only global biomedical embeddings into decoder feature space and selectively integrates them through semantic-local fusion and spatial gating. To the best of our knowledge, ThyCLIPNet is among the first lightweight thyroid ultrasound segmentation frameworks to use BiomedCLIP's vision encoder alone for image-only global semantic guidance without text prompting. Experiments on TG3K, TN3K, DDTI, and PKTN achieve dice similarity coefficients of 96.22%, 87.58%, 84.73%, and 80.70%; intersection over union scores of 92.72%, 77.91%, 73.51%, and 67.64%; and 95th-percentile hausdorff distances of 3.75, 16.38, 18.23, and 10.86, respectively. ThyCLIPNet uses 8.55M parameters and 22.99G FLOPs. Overall, the results support integrating global biomedical semantic guidance with lightweight multi-scale CNN representations for robust and computationally efficient thyroid ultrasound segmentation. Source code: https://github.com/Tasnim-Jahan/ThyCLIPNet. [Abstract shortened for arXiv. See PDF for full abstract.]