University Hospital Tuebingen, Department of Radiology, University of Tuebingen, Hoppe-Seyler-Straße 3 Tuebingen 72076, Germany; University of Tuebingen, Institute for Visual Computing, Department of Computer Science, Sand 14 Tuebingen 72076, Germany.
University Hospital Tuebingen, Department of Radiology, University of Tuebingen, Hoppe-Seyler-Straße 3 Tuebingen 72076, Germany.
Comput Methods Programs Biomed. 2022 Oct;225:107085. doi: 10.1016/j.cmpb.2022.107085. Epub 2022 Aug 27.
Tracking of anatomical structures in time-resolved medical image data plays an important role for various tasks such as volume change estimation or treatment planning. State-of-the-art deep learning techniques for automated tracking, while providing accurate results, require large amounts of human-labeled training data making their wide-spread use time- and resource-intensive. Our contribution in this work is the implementation and adaption of a self-supervised learning (SSL) framework that addresses this bottleneck of training data generation.
To this end we adapted and implemented an SSL framework that allows for automated anatomical tracking without the necessity for human-labeled training data. We evaluated this method by comparison to conventional- and deep learning optical flow (OF)-based tracking methods. We applied all methods on three different time-resolved medical image datasets (abdominal MRI, cardiac MRI, and echocardiography) and assessed their accuracy regarding tracking of pre-defined anatomical structures within and across individuals.
We found that SSL-based tracking as well as OF-based methods provide accurate results for simple, rigid and smooth motion patterns. However, regarding more complex motion, e.g. non-rigid or discontinuous motion patterns in the cardiac region, and for cross-subject anatomical matching, SSL-based tracking showed markedly superior performance.
We conclude that automated tracking of anatomical structures on time-resolved medical image data with minimal human labeling effort is feasible using SSL and can provide superior results compared to conventional and deep learning OF-based methods.
在时变医学图像数据中跟踪解剖结构对于各种任务（如体积变化估计或治疗计划）非常重要。用于自动跟踪的最先进的深度学习技术虽然提供了准确的结果，但需要大量的人工标记训练数据，因此广泛使用会耗费时间和资源。我们在这项工作中的贡献是实现和改编了一种自监督学习（SSL）框架，该框架可以解决训练数据生成的这一瓶颈。
为此，我们改编和实现了一种 SSL 框架，该框架允许在不需要人工标记训练数据的情况下进行自动解剖跟踪。我们通过与传统的和基于深度学习光流（OF）的跟踪方法进行比较来评估该方法。我们将所有方法应用于三个不同的时变医学图像数据集（腹部 MRI、心脏 MRI 和超声心动图），并评估了它们在个体内和个体间跟踪预定义解剖结构的准确性。
我们发现基于 SSL 的跟踪以及基于 OF 的方法对于简单、刚性和平滑的运动模式提供了准确的结果。然而，对于更复杂的运动，例如心脏区域的非刚性或不连续运动模式，以及跨个体的解剖匹配，基于 SSL 的跟踪表现出明显更好的性能。
我们得出结论，使用 SSL 可以最小的人工标记工作量对时变医学图像数据中的解剖结构进行自动跟踪，并且与传统的和基于深度学习 OF 的方法相比，它可以提供更好的结果。
