基于深度学习的超声诊断模型用于滤泡性甲状腺癌
这项回顾性多中心研究纳入10771例接受常规超声和术后病理的成人患者,开发端到端深度学习模型将甲状腺肿瘤分类为良性、滤泡性甲状腺癌和其他恶性肿瘤。在1078例测试集中,模型区分滤泡性甲状腺癌与其他恶性亚型及良性肿瘤的平衡准确率0.87、AUC 0.96(95%CI:0.96,0.96)、平均敏感度0.87、平均特异度0.92,优于参与研究的放射科医生(平衡准确率:初级0.60,P<0.001;中级0.59,P<0.001;高级0.66,P<0.001)。
要点
- 模型在测试集区分滤泡性甲状腺癌与其他恶性亚型及良性肿瘤的AUC为0.96(95%CI:0.96,0.96),平衡准确率0.87。
- 测试集平均敏感度0.87、平均特异度0.92,滤泡性甲状腺癌占3.6%(39/1078)。
- 模型平衡准确率优于初级(0.60)、中级(0.59)和高级(0.66)放射科医生,P均<0.001。
核心结论
- 所开发的分类模型解决了类别不平衡问题,在区分滤泡性甲状腺癌与其他恶性亚型及良性肿瘤方面较现有方法取得更高性能。
研究设计与样本
- 回顾性多中心研究,纳入2018年1月至2021年9月期间接受常规超声和术后病理的10771例连续成人患者;采用新型数据增强方法和混合损失函数处理不平衡数据集,应用于预训练的卷积神经网络和Transformer模型;测试集1078例(平均年龄47.3岁±11.8(标准差);女性811例;滤泡性甲状腺癌39/1078(3.6%);其他恶性肿瘤385/1078(35.7%))。
诊断标准与关键数值
- 模型在测试集区分滤泡性甲状腺癌与其他恶性亚型及良性肿瘤的平衡准确率0.87、AUC 0.96(95%CI:0.96,0.96)、平均敏感度0.87、平均特异度0.92;放射科医生平衡准确率:初级0.60(P<0.001)、中级0.59(P<0.001)、高级0.66(P<0.001)。
临床应用与阅片要点
- 作者认为该模型是有前景的计算机辅助诊断工具,用于滤泡性甲状腺癌的临床诊断,可能有助于减少滤泡性甲状腺癌的漏诊和误诊。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVES: It is challenging to preoperatively diagnose follicular thyroid carcinoma (FTC) on ultrasound images. This study aimed to develop an end-to-end diagnostic model that can classify thyroid tumors into benign tumors, FTC and other malignant tumors based on deep learning.
MATERIALS AND METHODS: This retrospective multi-center study included 10,771 consecutive adult patients who underwent conventional ultrasound and postoperative pathology between January 2018 and September 2021. We proposed a novel data augmentation method and a mixed loss function to solve an imbalanced dataset and applied them to a pre-trained convolutional neural network and transformer model that could effectively extract image features. The proposed model can directly identify FTC from other malignant subtypes and benign tumors based on ultrasound images.
RESULTS: The testing dataset included 1078 patients (mean age, 47.3 years ± 11.8 (SD); 811 female patients; FTCs, 39 of 1078 (3.6%); Other malignancies, 385 of 1078 (35.7%)). The proposed classification model outperformed state-of-the-art models on differentiation of FTC from other malignant sub-types and benign ones, achieved an excellent diagnosis performance with balanced-accuracy 0.87, AUC 0.96 (95% CI: 0.96, 0.96), mean sensitivity 0.87 and mean specificity 0.92. Meanwhile, it was superior to radiologists included in this study for thyroid tumor diagnosis (balanced-accuracy: Junior 0.60, p < 0.001; Mid-level 0.59, p < 0.001; Senior 0.66, p < 0.001).
CONCLUSION: The developed classification model addressed the class-imbalanced problem and achieved higher performance in differentiating FTC from other malignant subtypes and benign tumors compared with existing methods.
KEY POINTS: Question Deep learning has the potential to improve preoperatively diagnostic accuracy for follicular thyroid carcinoma (FTC). Findings The proposed model achieved high accuracy, sensitivity and specificity in diagnosing follicular thyroid carcinoma, outperforming other models. Clinical relevance The proposed model is a promising computer-aided diagnostic tool for the clinical diagnosis of FTC, which potentially could help reduce missed diagnosis and misdiagnosis for FTC.
引用
- Wang Y, Lu W, Xu L, et al. Deep learning-based ultrasound diagnostic model for follicular thyroid carcinoma. Eur Radiol. 2025. PMID: 40681777. doi:10.1007/s00330-025-11840-6
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。