智能体无代码AI工作流开发甲状腺结节超声恶性分类器
本研究测试自主无代码AI智能体能否开发校准良好的甲状腺结节恶性分类器,并在内部及外部多组织学癌症队列中验证。智能体使用开源数据集自主训练和校准模型,内部测试AUROC为0.94,外部验证AUROC为0.90,优于已有基准模型,但临床使用前需前瞻性验证和本地重校准。
要点
- 无代码智能体可自主开发校准良好的甲状腺结节恶性分类器。
- 外部验证AUROC为0.90,阴性预测值0.96。
- 漏诊癌症集中于滤泡型肿瘤。
核心结论
- 智能体无代码AI工作流可产生校准且外部验证的甲状腺结节分类器,支持可及、可重复和可独立测试的医学AI开发。
研究设计与样本
- 回顾性计算诊断研究,使用开源TN5000数据集,由无代码智能体自主审计数据、选择训练模型并校准概率。
- 锁定ResNet-18模型在科罗拉多大学232个结节上进行外部验证,并与既往发表分类器在同一来源上比较。
主要发现
- 内部测试集AUROC为0.94(95% CI,0.92-0.95),灵敏度0.90,特异度0.81。
- 外部验证AUROC为0.90(95% CI,0.85-0.93),灵敏度0.92,特异度0.68,阳性预测值0.52,阴性预测值0.96。
- 外部验证性能超过基准模型(AUROC 0.83;配对DeLong P = .03)。
- 漏诊癌症集中于滤泡型肿瘤。
临床怎么用
- 该无代码智能体工作流为甲状腺结节超声恶性分类器的开发提供了可及、可重复且可独立测试的途径。
- 临床使用前需进行前瞻性验证和本地重校准。
局限性
- 需前瞻性验证和本地重校准后方可用于临床。
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)。
原文
BACKGROUND: Convolutional neural networks can classify thyroid nodules on ultrasound, yet published models are seldom available for independent testing, require machine learning expertise to develop and deploy, and are validated mostly on papillary thyroid carcinoma. OBJECTIVE: To test whether an autonomous ("agentic"), no-code artificial intelligence (AI) agent can develop a calibrated thyroid nodule malignancy classifier and validate it internally and on an external cohort spanning multiple cancer histologies. METHODS: This was a retrospective, computational diagnostic study. A no-code agent (Hugging Face ML-Intern) autonomously audited the data, selected and trained the model, and calibrated probabilities, using the open-source TN5000 dataset. The locked ResNet-18 model was externally validated on 232 nodules from the University of Colorado and benchmarked against a previously published classifier evaluated on the same source. RESULTS: On the internal test set, the agentic model achieved an area under the receiver-operating-characteristic curve (AUROC) of 0.94 (95% CI, 0.92-0.95), sensitivity 0.90, and specificity 0.81. On external validation, the model achieved an AUROC of 0.90 (95% CI, 0.85-0.93); sensitivity was 0.92, specificity 0.68, positive predictive value 0.52, and negative predictive value 0.96, exceeding the performance of the benchmark model (AUROC 0.83; paired DeLong P = .03). Missed cancers clustered among follicular pattern tumors. CONCLUSION: An agentic, no-code AI workflow produced a calibrated, externally validated thyroid nodule classifier, supporting accessible, reproducible, and independently testable medical AI development. Prospective validation and local recalibration are required before clinical use.
引用
- Thomas J, Pozdeyev N. An agentic, no-code artificial intelligence workflow for developing a thyroid nodule ultrasound malignancy classifier. J Clin Endocrinol Metab. 2026. PMID: 42808848. doi:10.1210/clinem/dgag351
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。