Automated human brain tissue segmentation from clinical MRI images for improved neurological diagnosis and treatment planning

Abstract

ObjectiveSegmentation of medical images is a crucial process in various image analysis applications. Automated segmentation methods excel in accuracy compared to manual segmentation in the context of medical image analysis. One of the essential phases in the quantitative analysis of the brain is automated brain tissue segmentation using clinically obtained magnetic resonance imaging (MRI) data. It allows for precise quantitative examination of the brain, which aids in diagnosing, identifying, and classifying disorders. As a result, the segmentation approach's efficacy is crucial to disease diagnosis and treatment planning. MethodsThis paper presents a hybrid optimization method for segmenting brain tissue from clinical MRI scans using a fractional Henry horse herd gas optimization-based Shepherd convolutional neural network (FrHHGO-based ShCNN). To segment the clinical brain MRI images into white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF) tissues, the proposed framework was evaluated on the Lifespan Human Connectome Projects (HCP) database. The hybrid optimization algorithm, FrHHGO, integrates the fractional Henry gas optimization (FHGO) and horse herd optimization (HHO) algorithms. Training required 30 minutes, whereas testing and segmenting brain tissues from an unseen image took an average of 12 seconds. ResultsCompared to the results obtained with no refinement, the Skull stripping refinement showed a significant improvement. Since the method includes a preprocessing stage, it is flexible enough to enhance image quality, allowing for better results even with low-resolution input. Maximum precision of 93.2%, recall of 91.5%, dice score of 91.1% and F1-score of 90.5% was achieved by the suggested FrHHGO-based ShCNN, which was superior to all other approaches. ConclusionThe proposed method may outperform existing state-of-the-art methodologies in both qualitative and quantitative measurements across a wide range of medical modalities. It could demonstrate its potential for real-life clinical application.