Abstract
Segmentation is crucial in medical imaging analysis to help extract regions of interest (ROI) from different imaging modalities. The aim of this study is to develop and train a 3D convolutional neural network (CNN) for skull segmentation in magnetic resonance imaging (MRI). 58 gold standard volumetric labels were created from computed tomography (CT) scans in standard tessellation language (STL) models. These STL models were converted into matrices and overlapped on the 58 corresponding MR images to create the MRI gold standards labels. The CNN was trained with these 58 MR images and a mean ± standard deviation (SD) Dice similarity coefficient (DSC) of 0.7300 ± 0.04 was achieved. A further investigation was carried out where the brain region was removed from the image with the help of a 3D CNN and manual corrections by using only MR images. This new dataset, without the brain, was presented to the previous CNN which reached a new mean ± SD DSC of 0.7826 ± 0.03. This paper aims to provide a framework for segmenting the skull using CNN and STL models, as the 3D CNN was able to segment the skull with a certain precision.
Highlights
Image segmentation is the process of partitioning an image into multiple sections to simplify the image into something more meaningful so that we can locate regions of interest (ROI)
We used the cancer imaging archive data collections (TCIA) [30] to search for reliable datasets that contain computed tomography (CT) and Magnetic resonance imaging (MRI) from the same patient and a minimum variation in the coronal, sagittal, and transverse plane. 58 volumetric CT and MR images were selected from four datasets to meet these criteria:
The results found in this article reflect a long-standing search for the development of a technique for bone segmentation in MRI, the proposed method Dice similarity coefficient (DSC) (0.7826 ± 0.03)
Summary
Image segmentation is the process of partitioning an image into multiple sections to simplify the image into something more meaningful so that we can locate regions of interest (ROI). In medical imaging and analysis, these ROI, identified by the segmentation process in an image scanning system, can represent various structures in the body such as pathologies, tissues, bone, organs, prosthesis, and so forth. Magnetic resonance imaging (MRI) and computed tomography (CT) are the most common medical image scanning system used to reveal relevant structures for automated processing of scanned data. Both techniques are excellent in providing non-invasive diagnostic images of organs and structures inside the body. A study in [1] pointed out that the risk to develop leukemia and brain tumors increases with the radiation exposure from CT scans.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have