Combined Atlas and Convolutional Neural Network-Based Segmentation of the Hippocampus from MRI According to the ADNI Harmonized Protocol.

Hippocampus atrophy is an early structural feature that can be measured from magnetic resonance imaging (MRI) to improve the diagnosis of neurological diseases. An accurate and robust standardized hippocampus segmentation method is required for reliable atrophy assessment. The aim of this work was to develop and evaluate an automatic segmentation tool (DeepHarp) for hippocampus delineation according to the ADNI harmonized hippocampal protocol (HarP). DeepHarp utilizes a two-step process. First, the approximate location of the hippocampus is identified in T1-weighted MRI datasets using an atlas-based approach, which is used to crop the images to a region-of-interest (ROI) containing the hippocampus. In the second step, a convolutional neural network trained using datasets with corresponding manual hippocampus annotations is used to segment the hippocampus from the cropped ROI. The proposed method was developed and validated using 107 datasets with manually segmented hippocampi according to the ADNI-HarP standard as well as 114 multi-center datasets of patients with Alzheimer's disease, mild cognitive impairment, cerebrovascular disease, and healthy controls. Twenty-three independent datasets manually segmented according to the ADNI-HarP protocol were used for testing to assess the accuracy, while an independent test-retest dataset was used to assess precision. The proposed DeepHarp method achieved a mean Dice similarity score of 0.88, which was significantly better than four other established hippocampus segmentation methods used for comparison. At the same time, the proposed method also achieved a high test-retest precision (mean Dice score: 0.95). In conclusion, DeepHarp can automatically segment the hippocampus from T1-weighted MRI datasets according to the ADNI-HarP protocol with high accuracy and robustness, which can aid atrophy measurements in a variety of pathologies.

Case Study: Intra-Patient Heterogeneity of Aneurysmal Tissue Properties.

Introduction: Current recommendations for surgical treatment of abdominal aortic aneurysms (AAAs) rely on the assessment of aortic diameter as a marker for risk of rupture. The use of aortic size alone may overlook the role that vessel heterogeneity plays in aneurysmal progression and rupture risk. The aim of the current study was to investigate intra-patient heterogeneity of mechanical and fluid mechanical stresses on the aortic wall and wall tissue histopathology from tissue collected at the time of surgical repair. Methods: Finite element analysis (FEA) and computational fluid dynamics (CFD) simulations were used to predict the mechanical wall stress and the wall shear stress fields for a non-ruptured aneurysm 2 weeks prior to scheduled surgery. During open repair surgery one specimen partitioned into different regions was collected from the patient's diseased aorta according to a pre-operative map. Histological analysis and mechanical testing were performed on the aortic samples and the results were compared with the predicted stresses. Results: The preoperative simulations highlighted the presence of altered local hemodynamics particularly at the proximal segment of the left anterior area of the aneurysm. Results from the post-operative assessment on the surgical samples revealed a considerable heterogeneity throughout the aortic wall. There was a positive correlation between elastin fragmentation and collagen content in the media. The tensile tests demonstrated a good prediction of the locally varying constitutive model properties predicted using geometrical variables, i.e., wall thickness and thrombus thickness. Conclusions: The observed large regional differences highlight the local response of the tissue to both mechanical and biological factors. Aortic size alone appears to be insufficient to characterize the large degree of heterogeneity in the aneurysmal wall. Local assessment of wall vulnerability may provide better risk of rupture predictions.