Skull stripping is a key step in the pre-processing of rodent brain magnetic resonance images (MRI). This study aimed to develop a new skull stripping method via U-Net, a neural network model based on deep learning method, for rat brain MRI.In this study, 599 rats were enrolled and U-Net was applied to segment MRI images of rat brain. The intercranial tissue of each rat was manually labeled. 476 rats (approximate 80%) were used for training set while 123 rats (approximate 20%) were used to test the performance of the trained U-Net model. For evaluation, the segmentation result by the U-Net model is compared with the manual label, and traditional segment methods. Quantitative evaluation, including Dice coefficient, Jaccard coefficient, Sensitivity, Specificity, Pixel accuracy, Hausdorff coefficient, True positive rate, False positive rate and the volumes of whole brain, were calculated to compare the segmentation results among different models.The U-Net model was performed better than the software of RATS and BrainSuite, in which the quantitative values of training U-Net model were 0.9907 ± 0.0016 (Dice coefficient), 0.9816 ± 0.0032 (Jaccard coefficient), 0.9912 ± 0.0020 (Sensitivity), 0.9989 ± 0.0002 (Specificity), 0.9982 ± 0.0003 (Pixel accuracy), 5.2390 ± 2.5334 (Hausdorff coefficient), 0.9902 ± 0.0025 (True positive rate), 0.0009 ± 0.0002(False positive rate) respectively.This study provides a new method that achieves reliable performance in rat brain skull stripping of MRI images, which could contribute to the processing of rat brain MRI.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1088/1361-6560/acf641 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!