Department of Bioengineering, The University of Texas at Dallas, Richardson, Texas, USA.
Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, Texas, USA.
Med Phys. 2022 Feb;49(2):1153-1160. doi: 10.1002/mp.15404. Epub 2021 Dec 22.
The goal is to study the performance improvement of a deep learning algorithm in three-dimensional (3D) image segmentation through incorporating minimal user interaction into a fully convolutional neural network (CNN).
A U-Net CNN was trained and tested for 3D prostate segmentation in computed tomography (CT) images. To improve the segmentation accuracy, the CNN's input images were annotated with a set of border landmarks to supervise the network for segmenting the prostate. The network was trained and tested again with annotated images after 5, 10, 15, 20, or 30 landmark points were used.
Compared to fully automatic segmentation, the Dice similarity coefficient increased up to 9% when 5-30 sparse landmark points were involved, with the segmentation accuracy improving as more border landmarks were used.
When a limited number of sparse border landmarks are used on the input image, the CNN performance approaches the interexpert observer difference observed in manual segmentation.
通过将最小限度的用户交互纳入全卷积神经网络（CNN），研究深度学习算法在三维（3D）图像分割方面的性能提升。
在 CT 图像中，使用 U-Net CNN 对前列腺进行 3D 分割。为了提高分割准确性，用一组边界标记点对 CNN 的输入图像进行注释，以指导网络对前列腺进行分割。在使用 5、10、15、20 或 30 个标记点后，使用带注释的图像再次对网络进行训练和测试。
与全自动分割相比，当涉及 5-30 个稀疏标记点时，Dice 相似系数提高了高达 9%，并且随着使用更多的边界标记点，分割准确性也得到了提高。
当输入图像上使用有限数量的稀疏边界标记点时，CNN 的性能接近手动分割中观察到的专家间观察者差异。
