Automatic segmentation of brain tumor in intraoperative ultrasound images using 3D U-Net

Abstract

Because of the deformation of the brain during neurosurgery, intraoperative imaging can be used to visualize the actual location of the brain structures. These images are used for image-guided navigation as well as determining whether the resection is complete and localizing the remaining tumor tissue. Intraoperative ultrasound (iUS) is a convenient modality with short acquisition times. However, iUS images are difficult to interpret because of the noise and artifacts. In particular, tumor tissue is difficult to distinguish from healthy tissue and it is very difficult to delimit tumors in iUS images. In this paper, we propose an automatic method to segment low grade brain tumors in iUS images using a 2-D and 3-D U-Net. We trained the networks on three folds with twelve training cases and five test cases each. The obtained results are promising, with a median Dice score of 0.72. The volume differences between the estimated and ground truth segmentations were similar to the intra-rater volume differences. While these results are preliminary, they suggest that deep learning methods can be successfully applied to tumor segmentation in intraoperative images.

Multiclass segmentation of brain intraoperative ultrasound images with limited data

“…16 The tumor segmentations were created for a study on tumor volume 20 and were also used in a single class segmentation model. 15 Sulci , falx cerebri and ventricles were segmented for this study.…”

“…Segmenting structures in medical images has been widely studied, for which deep learning is the current state-of-the-art. 11 This has been applied to ultrasound images of the brain, to segment the midbrain, 12 sulci and falx cerebri , 13,14 tumor, 15 resection cavity. 16 However, multi-class segmentation models have been shown to obtain better results in other applications, 17,18 leveraging inter-class dependencies.…”