Improved Brain Segmentation Using Pixel Separation And Additional Segmentation Features

Brain segmentation is vital for brain structure evaluation of disease diagnosis and treatment.Many research has been conducted to study brain segmentation.However,prior research has not considered separating actual brain pixels from the background of brain images.Not performing such separation may(a)distort brain segmentation models and(b)introduce overhead to the modeling performance.In this paper,we improve the performance of brain segmentation using 3D,fully Convolutional Neural Network(CNN)models.The Infant and adult datasets,and a multi-instance loss method are used to separate actual brain pixels from the background,and Gabor filter banks and K-means clustering are applied to provide additional segmentation features.Our model reached dice coefficients of 87.4%–94.1%(i.e.,an improvement of up to 11% to the results of five stateof-the-art models).Unlike prior studies,experts in medical imaging are invited to evaluate the segmentation results.We observe that our results are fairly close to the manual reference.Besides,our model preforms 1.2x–2.6x faster than previous models.Which concludes that our model is more efficient and accurate in practice for both infant and adult brain segmentation.