The study population consisted of 73 patients. All patients had an aneurysm of the ascending aorta, and surgery was required. Patients with connective tissue disease were not included in our study group. MRI acquisition was performed pre-operatively. The acquisition protocol included a sequence of images specifically for the evaluation of aortic compliance. On average, 30 slices of cine-MRI were acquired for each patient. The patient’s blood pressure was recorded during the exam and used to calculate the aortic compliance.
A deep learning model was trained to segment the ascending aorta automatically. The aortic compliance was calculated based on the segmented aorta. In parallel, for comparison, a medical expert performed semi-automatic contouring of the ascending aorta using the commercial software QIR (CASIS, Quetigny, France). Using the center of gravity of the aorta, we automatically divided the aortic surface into four quadrants to calculate the local elastic properties.
For the ex-vivo evaluation, the collected aortic wall samples from the replacement procedure were partitioned relative to medial, posterior, lateral, and anterior quadrants. The quadrants of the aorta were stretched biaxially until rupture and Young’s modulus was calculated. This parameter measures the ability of a material to withstand changes in length (strain) when under lengthwise tension or compression (tensile stress).
We compared the results of the in-vivo and ex-vivo experiments, leveraging common patients between the datasets. The compared values are in-vivo compliance, strain, and ex-vivo Young’s modulus and strain.
A deep learning model was trained to segment the ascending aorta automatically. The aortic compliance was calculated based on the segmented aorta. In parallel, for comparison, a medical expert performed semi-automatic contouring of the ascending aorta using the commercial software QIR (CASIS, Quetigny, France). Using the center of gravity of the aorta, we automatically divided the aortic surface into four quadrants to calculate the local elastic properties.
For the ex-vivo evaluation, the collected aortic wall samples from the replacement procedure were partitioned relative to medial, posterior, lateral, and anterior quadrants. The quadrants of the aorta were stretched biaxially until rupture and Young’s modulus was calculated. This parameter measures the ability of a material to withstand changes in length (strain) when under lengthwise tension or compression (tensile stress).
We compared the results of the in-vivo and ex-vivo experiments, leveraging common patients between the datasets. The compared values are in-vivo compliance, strain, and ex-vivo Young’s modulus and strain.
Free full text: Click here
