CMCNet: Aligning Ultrasound Image Embeddings with Textual TI-RADS Representations for Fine-Grained Thyroid Classification

2026-08-14

AI Summary

This study addresses the underutilization of feature-level supervision in multi-class thyroid nodule ultrasound classification by proposing CMCNet. The method employs text embeddings as stable proxy representations for risk stratification, achieving cross-modal alignment between ultrasound images and TI-RADS textual descriptions via a Center-Margin Contrastive loss to guide image-only fine-grained grading with structured features. Experimental results demonstrate that CMCNet significantly outperforms InfoNCE and multi-task baselines in data efficiency and robustness. Notably, it exhibits superior performance under class-imbalanced conditions, effectively enhancing the accuracy of nodule risk stratification.

Abstract

Ultrasound is the primary imaging modality for assessing thyroid nodules, and the ACR TI-RADS framework standardizes diagnosis through five ultrasound feature categories that are aggregated into five risk levels (TR1-TR5). Although widely adopted in clinical practice, most deep learning approaches focus on binary malignancy classification, while multi-class prediction and explicit utilization of feature-level supervision remain underexplored, largely due to limited annotated data. In this study, we introduce the STN dataset of 600 thyroid nodules with paired transverse and longitudinal ultrasound images, bounding box annotations, and complete labels for all five TI-RADS feature categories. Following the clinical decision process, we investigate how structured feature information can guide representation learning during training while requiring only images at inference. We demonstrate that text embeddings derived from standardized feature descriptions form a stable surrogate representation for TI-RADS risk levels. Based on this observation, we propose CMCNet, which aligns image embeddings to fixed textual embeddings via a Center-Margin Contrastive Loss that simultaneously promotes intra-class compactness and inter-class separation. Experimental results show that this embedding alignment strategy is more data-efficient and robust than direct multitask learning, and consistently outperforms InfoNCE, center loss, a strong multitask baseline, and a VQA-style multimodal model, particularly in imbalanced settings. The dataset is freely available at doi: 10.5281/zenodo.19125693 and the source code is available at: https://www.healthinformaticslab.org/supp/.

Problem

• Thyroid Nodule Classification
• TI-RADS
• Fine-Grained Classification
• Ultrasound Imaging
• Limited Annotated Data

Innovation

• Center-Margin Contrastive Loss
• Image-Text Alignment
• TI-RADS
• Thyroid Nodule Classification
• Representation Learning