Artificial Intelligence in Biomedical Imaging Lab, University of Pennsylvania, Philadelphia, Pennsylvania, USA.
Biggs Alzheimer's Institute, University of Texas San Antonio Health Science Center, San Antonio, Texas, USA.
J Magn Reson Imaging. 2022 Mar;55(3):908-916. doi: 10.1002/jmri.27908. Epub 2021 Sep 25.
In the medical imaging domain, deep learning-based methods have yet to see widespread clinical adoption, in part due to limited generalization performance across different imaging devices and acquisition protocols. The deviation between estimated brain age and biological age is an established biomarker of brain health and such models may benefit from increased cross-site generalizability.
To develop and evaluate a deep learning-based image harmonization method to improve cross-site generalizability of deep learning age prediction.
Retrospective.
Eight thousand eight hundred and seventy-six subjects from six sites. Harmonization models were trained using all subjects. Age prediction models were trained using 2739 subjects from a single site and tested using the remaining 6137 subjects from various other sites.
FIELD STRENGTH/SEQUENCE: Brain imaging with magnetization prepared rapid acquisition with gradient echo or spoiled gradient echo sequences at 1.5 T and 3 T.
StarGAN v2, was used to perform a canonical mapping from diverse datasets to a reference domain to reduce site-based variation while preserving semantic information. Generalization performance of deep learning age prediction was evaluated using harmonized, histogram matched, and unharmonized data.
Mean absolute error (MAE) and Pearson correlation between estimated age and biological age quantified the performance of the age prediction model.
Our results indicated a substantial improvement in age prediction in out-of-sample data, with the overall MAE improving from 15.81 (±0.21) years to 11.86 (±0.11) with histogram matching to 7.21 (±0.22) years with generative adversarial network (GAN)-based harmonization. In the multisite case, across the 5 out-of-sample sites, MAE improved from 9.78 (±6.69) years to 7.74 (±3.03) years with histogram normalization to 5.32 (±4.07) years with GAN-based harmonization.
While further research is needed, GAN-based medical image harmonization appears to be a promising tool for improving cross-site deep learning generalization.
4 TECHNICAL EFFICACY: Stage 1.
在医学成像领域，基于深度学习的方法尚未得到广泛的临床应用，部分原因是在不同的成像设备和采集协议之间的泛化性能有限。估计脑龄与生物年龄的偏差是大脑健康的一个既定生物标志物，此类模型可能受益于提高跨站点的可推广性。
开发并评估一种基于深度学习的图像调和方法，以提高深度学习年龄预测的跨站点泛化能力。
回顾性。
来自六个地点的 8876 名受试者。使用所有受试者进行调和模型训练。使用来自单个站点的 2739 名受试者训练年龄预测模型，并使用来自其他各种站点的 6137 名剩余受试者进行测试。
磁场强度/序列：在 1.5T 和 3T 下使用磁化准备快速获取梯度回波或扰相梯度回波序列进行脑成像。
使用 StarGAN v2 对来自不同数据集的图像进行规范映射，将其映射到参考域，以减少基于站点的变化，同时保留语义信息。使用调和、直方图匹配和未调和的数据评估深度学习年龄预测的泛化性能。
估计年龄与生物年龄之间的平均绝对误差（MAE）和 Pearson 相关系数用于量化年龄预测模型的性能。
我们的结果表明，在样本外数据中，年龄预测得到了实质性的提高，整体 MAE 从 15.81（±0.21）岁提高到 11.86（±0.11）岁，直方图匹配提高到 7.21（±0.22）岁，生成对抗网络（GAN）基于调和的。在多站点情况下，在 5 个样本外站点中，MAE 从 9.78（±6.69）岁提高到 7.74（±3.03）岁，直方图归一化提高到 5.32（±4.07）岁，GAN 基于调和。
虽然需要进一步研究，但基于 GAN 的医学图像调和似乎是提高跨站点深度学习推广的有前途的工具。
4 技术功效：阶段 1。
