A Study of Ultrasound Image Segmentation of Thyroid Nodules Based on Feature Adaptive Extraction and Gated Fusion Mechanisms

Objective To address the problems of high heterogeneity of nodule regions, blurred boundaries and noise interference in thyroid ultrasound images, a deep learning network FAGF-Net based on feature adaptive extraction and gated fusion mechanism is proposed to improve the segmentation accuracy of thyroid nodules.Methods A feature coupling encoder is designed to integrate CNN and Transformer in parallel to capture local texture details and global structural semantics, respectively; a multi-scale feature spatial coupling module is proposed to enhance the capture of multi-scale geometric features of the nodule by using dynamic multi-scale convolutional kernel and the channel-spatial attention mechanism; and a contextual gating feature attention module is proposed to filter the multi-level redundant features through the gating mechanism, and enhance the cross-level redundancy with the self-attention mechanism; the FAGF-Net is designed to enhance the segmentation accuracy of thyroid nodule by using the self-attention mechanism. We use the context-gated feature attention module to filter multi-layer redundant features through the gating mechanism, and enhance cross-layer feature associations with the help of self-attention to suppress noise interference.Results The TN3K dataset is used in the study, and the results show that the accuracy of FAGF-Net reaches 95.59%, which is 3.77% higher than that of the baseline model UNet; the Dice coefficient is 90.60%, which is 3.55% higher. The ablation experiments show that FC-Encoder, MFSE-Module and CGFA-Module improve the Dice coefficient by 1.40%, 0.69% and 2.41%, respectively, which verifies the necessity of each module. The visualisation results show that FAGF-Net significantly outperforms the comparison models in terms of boundary fitting smoothness, accuracy of tiny lesion localisation, and effectively reduces the phenomena of under-segmentation and over-segmentation.Conclusion FAGF-Net significantly improves the segmentation performance of ultrasound images of thyroid nodules through the innovative design of feature coupling and gating mechanism, providing a new solution for clinical automated diagnosis.