Candemir Sema, Nguyen Xuan V, Folio Les R, Prevedello Luciano M
Department of Radiology, The Ohio State University College of Medicine, 395 W 12th Ave, Columbus, OH 43212 (S.C., X.V.N., L.M.P.); and Department of Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, Md (L.R.F.).
Radiol Artif Intell. 2021 Oct 6;3(6):e210014. doi: 10.1148/ryai.2021210014. eCollection 2021 Nov.
Data-driven approaches have great potential to shape future practices in radiology. The most straightforward strategy to obtain clinically accurate models is to use large, well-curated and annotated datasets. However, patient privacy constraints, tedious annotation processes, and the limited availability of radiologists pose challenges to building such datasets. This review details model training strategies in scenarios with limited data, insufficiently labeled data, and/or limited expert resources. This review discusses strategies to enlarge the data sample, decrease the time burden of manual supervised labeling, adjust the neural network architecture to improve model performance, apply semisupervised approaches, and leverage efficiencies from pretrained models. Computer-aided Detection/Diagnosis, Transfer Learning, Limited Annotated Data, Augmentation, Synthetic Data, Semisupervised Learning, Federated Learning, Few-Shot Learning, Class Imbalance.
数据驱动的方法在塑造放射学未来实践方面具有巨大潜力。获得临床准确模型的最直接策略是使用大型、精心策划和注释的数据集。然而，患者隐私限制、繁琐的注释过程以及放射科医生数量有限，给构建此类数据集带来了挑战。本综述详细介绍了在数据有限、标注数据不足和/或专家资源有限的情况下的模型训练策略。本综述讨论了扩大数据样本、减轻人工监督标注的时间负担、调整神经网络架构以提高模型性能、应用半监督方法以及利用预训练模型提高效率的策略。计算机辅助检测/诊断、迁移学习、有限标注数据、增强、合成数据、半监督学习、联邦学习、少样本学习、类别不平衡。
