文献速递骨显像与常规 SPECT心脏核医学

生成式人工智能实现骨显像图像生成并提升深度学习模型在数据受限环境中的泛化能力

Generative artificial intelligence enables the generation of bone scintigraphy images and improves generalization of deep learning models in data-constrained environments.
PubMed 文献 · Eur J Nucl Med Mol Imaging · 2025-01-29

本研究利用9,170例患者的99mTc-骨显像训练生成模型,合成骨转移与心脏淀粉样变性两种模式的匿名标注图像;四名读者无法区分合成与真实扫描(平均准确率0.48%[95%CI 0.46-0.51],400例中239例(60%)判断不一致,Fleiss kappa 0.18)。将合成数据加入小样本训练集后,骨转移异常摄取检测的AUC平均提升33(±10)%(p<0.0001),心脏淀粉样变性摄取检测的AUC平均提升5(±4)%(p<0.0001);预测为异常摄取的患者临床结局更差(log-rank p<0.0001)。

要点

  • 生成式AI可定向生成代表骨转移和心脏淀粉样变性两种疾病模式的骨显像图像。
  • 四名读者无法区分合成与真实扫描,平均准确率0.48%(95%CI 0.46-0.51)。
  • 加入合成数据使骨转移检测AUC平均提升33(±10)%,心脏淀粉样变性检测AUC平均提升5(±4)%。

核心结论

  • 生成式AI能够定向生成代表不同临床状况的骨显像图像。
  • 研究结果提示合成数据有潜力克服数据共享困难以及在数据有限环境中开发可靠且具预后价值的深度学习模型的挑战。

研究设计与样本

  • 研究在一个中心使用9,170例患者的99mTc-骨显像扫描训练生成模型,生成代表骨转移和心脏淀粉样变性两种疾病模式的高质量、完全匿名标注扫描。
  • 进行盲法读者研究评估生成数据的临床有效性和质量。
  • 用合成数据增强一个独立的小型单中心数据集,训练深度学习模型在下游分类任务中检测异常摄取,并在四个外部站点6,448例患者的7,472次扫描上以跨示踪剂和跨扫描仪设置测试模型,将预测与临床结局关联。

主要发现

  • 四名读者无法区分合成扫描与真实扫描,平均准确率为0.48%(95%CI 0.46-0.51),在400例中有239例(60%)判断不一致(Fleiss kappa:0.18)。
  • 与未使用合成数据训练的模型相比,加入合成数据使检测提示骨转移的异常摄取的AUC平均(±SD)提升33(±10)%(p<0.0001),使检测提示心脏淀粉样变性的摄取的AUC平均(±SD)提升5(±4)%(p<0.0001),涵盖内部和外部测试队列。
  • 预测为异常摄取的患者具有不良临床结局(log-rank:p<0.0001)。

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

PURPOSE: Advancements of deep learning in medical imaging are often constrained by the limited availability of large, annotated datasets, resulting in underperforming models when deployed under real-world conditions. This study investigated a generative artificial intelligence (AI) approach to create synthetic medical images taking the example of bone scintigraphy scans, to increase the data diversity of small-scale datasets for more effective model training and improved generalization. METHODS: We trained a generative model on 99mTc-bone scintigraphy scans from 9,170 patients in one center to generate high-quality and fully anonymized annotated scans of patients representing two distinct disease patterns: abnormal uptake indicative of (i) bone metastases and (ii) cardiac uptake indicative of cardiac amyloidosis. A blinded reader study was performed to assess the clinical validity and quality of the generated data. We investigated the added value of the generated data by augmenting an independent small single-center dataset with synthetic data and by training a deep learning model to detect abnormal uptake in a downstream classification task. We tested this model on 7,472 scans from 6,448 patients across four external sites in a cross-tracer and cross-scanner setting and associated the resulting model predictions with clinical outcomes. RESULTS: The clinical value and high quality of the synthetic imaging data were confirmed by four readers, who were unable to distinguish synthetic scans from real scans (average accuracy: 0.48% [95% CI 0.46-0.51]), disagreeing in 239 (60%) of 400 cases (Fleiss' kappa: 0.18). Adding synthetic data to the training set improved model performance by a mean (± SD) of 33(± 10)% AUC (p < 0.0001) for detecting abnormal uptake indicative of bone metastases and by 5(± 4)% AUC (p < 0.0001) for detecting uptake indicative of cardiac amyloidosis across both internal and external testing cohorts, compared to models without synthetic training data. Patients with predicted abnormal uptake had adverse clinical outcomes (log-rank: p < 0.0001). CONCLUSIONS: Generative AI enables the targeted generation of bone scintigraphy images representing different clinical conditions. Our findings point to the potential of synthetic data to overcome challenges in data sharing and in developing reliable and prognostic deep learning models in data-limited environments.

来源:Eur J Nucl Med Mol Imaging,查看原文。版权归原作者所有。

引用

  1. Haberl D, Ning J, Kluge K, et al. Generative artificial intelligence enables the generation of bone scintigraphy images and improves generalization of deep learning models in data-constrained environments. Eur J Nucl Med Mol Imaging. 2025. PMID: 39878897. doi:10.1007/s00259-025-07091-8

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。