Validation of a deformable MRI to CT registration algorithm employing same day planning MRI for surrogate analysis

Brain tumor segmentation based on region of interest-aided localization and segmentation U-Net

Since magnetic resonance imaging (MRI) has superior soft tissue contrast, contouring (brain) tumor accurately by MRI images is essential in medical image processing. Segmenting tumor accurately is immensely challenging, since tumor and normal tissues are often inextricably intertwined in the brain. It is also extremely time consuming manually. Late deep learning techniques start to show reasonable success in brain tumor segmentation automatically. The purpose of this study is to develop a new region-of-interest-aided (ROI-aided) deep learning technique for automatic brain tumor MRI segmentation. The method consists of two major steps. Step one is to use a 2D network with U-Net architecture to localize the tumor ROI, which is to reduce the impact of normal tissue’s disturbance. Then a 3D U-Net is performed in step 2 for tumor segmentation within identified ROI. The proposed method is validated on MICCAI BraTS 2015 Challenge with 220 high Gliomas grade (HGG) and 54 low Gliomas grade (LGG) patients’ data. The Dice similarity coefficient and the Hausdorff distance between the manual tumor contour and that segmented by the proposed method are 0.876 ±0.068 and 3.594±1.347 mm, respectively. These numbers are indications that our proposed method is an effective ROI-aided deep learning strategy for brain MRI tumor segmentation, and a valid and useful tool in medical image processing.

Localization of prostatic tumor's infection based on normalized mutual information MRI image segmentation

To investigate the effect of normalized mutual information (MRI) image segmentation in accurate localization of prostate cancer with infection and the role in disease treatment, the normalized mutual information method is used to measure the similarity of images, so as to select the maps. Then, the popular global weighted voting method and normalized mutual information method are applied to calculate the weights and carry out the label image fusion. The map selection method based on mutual information substantially completes the segmentation of the MRI image prostate. The prostate position is basically found on the 10 test images, and the positioning of the prostate organs is deviated in the worst case. In the case of poor multi-map segmentation, it usually happens when those are not well represented in the map. Because of the structural similarity of medical images, multi-atlas segmentation based on normalized mutual information method can be done. Using the prior information of atlas, the atlas label image can be selected. After fusion, the final segmentation of the test image can be completed, which has a high accuracy for the location of prostate cancer. This method can accurately delineate the target area in radiotherapy of prostate cancer and reduce the damage of rectum, bladder and other organs caused by radiotherapy. However, there are still some problems in this study, such as inadequate segmentation accuracy, long data processing time and so on. There is still a certain distance from practicality, and further research is needed.