Dataset Information

A Comparative Classification Analysis of Abdominal Aortic Aneurysms by Machine Learning Algorithms.

ABSTRACT: The objective of this work was to perform image-based classification of abdominal aortic aneurysms (AAA) based on their demographic, geometric, and biomechanical attributes. We retrospectively reviewed existing demographics and abdominal computed tomography angiography images of 100 asymptomatic and 50 symptomatic AAA patients who received an elective or emergent repair, respectively, within 1-6 months of their last follow up. An in-house script developed within the MATLAB computational platform was used to segment the clinical images, calculate 53 descriptors of AAA geometry, and generate volume meshes suitable for finite element analysis (FEA). Using a third party FEA solver, four biomechanical markers were calculated from the wall stress distributions. Eight machine learning algorithms (MLA) were used to develop classification models based on the discriminatory potential of the demographic, geometric, and biomechanical variables. The overall classification performance of the algorithms was assessed by the accuracy, area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and precision of their predictions. The generalized additive model (GAM) was found to have the highest accuracy (87%), AUC (89%), and sensitivity (78%), and the third highest specificity (92%), in classifying the individual AAA as either asymptomatic or symptomatic. The k-nearest neighbor classifier yielded the highest specificity (96%). GAM used seven markers (six geometric and one biomechanical) to develop the classifier. The maximum transverse dimension, the average wall thickness at the maximum diameter, and the spatially averaged wall stress were found to be the most influential markers in the classification analysis. A second classification analysis revealed that using maximum diameter alone results in a lower accuracy (79%) than using GAM with seven geometric and biomechanical markers. We infer from these results that biomechanical and geometric measures by themselves are not sufficient to discriminate adequately between population samples of asymptomatic and symptomatic AAA, whereas MLA offer a statistical approach to stratification of rupture risk by combining demographic, geometric, and biomechanical attributes of patient-specific AAA.

SUBMITTER: Rengarajan B

PROVIDER: S-EPMC7096253 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveTo identify confounding variables influencing the accuracy of a convolutional neural network (CNN) specific for infrarenal abdominal aortic aneurysms (AAAs) on computed tomography angiograms (CTAs).MethodsA Health Insurance Portability and Accountability Act-compliant, institutional review board-approved, retrospective study analyzed abdominopelvic CTA scans from 200 patients with infrarenal AAAs and 200 propensity-matched control patients. An AAA-specific trained CNN was developed by the application of transfer learning to the VGG-16 base model using model training, validation, and testing techniques. Model accuracy and area under the curve were analyzed based on data sets (selected, balanced, or unbalanced), aneurysm size, extra-abdominal extension, dissections, and mural thrombus. Misjudgments were analyzed by review of heatmaps, via gradient weighted class activation, overlaid on CTA images.ResultsThe trained custom CNN model reported high test group accuracies of 94.1%, 99.1%, and 99.6% and area under the curve of 0.9900, 0.9998, and 0.9993 in selected (n = 120), balanced (n = 3704), and unbalanced image sets (n = 31,899), respectively. Despite an eightfold difference between balanced and unbalanced image sets, the CNN model demonstrated high test group sensitivities (98.7% vs 98.9%) and specificities (99.7% vs 99.3%) in unbalanced and balanced image sets, respectively. For aneurysm size, the CNN model demonstrates decreasing misjudgments as aneurysm size increases: 47% (16/34) for aneurysms <3.3 cm, 32% (11/34) for aneurysms 3.3 to 5 cm, and 20% (7/34) for aneurysms >5 cm. Aneurysms containing measurable mural thrombus were over-represented within type II (false-negative) misjudgments compared with type I (false-positive) misjudgments (71% vs 15%, P < .05). Inclusion of extra-abdominal aneurysm extension (thoracic or iliac artery) or dissection flaps in these imaging sets did not decrease the model's overall accuracy, indicating that the model performance was excellent without the need to clean the data set of confounding or comorbid diagnoses.ConclusionsAnalysis of an AAA-specific CNN model can accurately screen and identify infrarenal AAAs on CTA despite varying pathology and quantitative data sets. The highest anatomic misjudgments were with small aneurysms (<3.3 cm) or the presence of mural thrombus. Accuracy of the CNN model is maintained despite the inclusion of extra-abdominal pathology and imbalanced data sets.

Project description:Abdominal aortic aneurysm (AAA) is an asymptomatic aortic disease with a survival rate of 20% after rupture. It is a vascular degenerative condition different from occlusive arterial diseases. The size of the aneurysm is the most important determining factor in its clinical management. However, other measures of the AAA geometry that are currently not used clinically may also influence its rupture risk. With this in mind, the objectives of this work are to develop an algorithm to calculate the AAA wall thickness and abdominal aortic diameter at planes orthogonal to the vessel centerline, and to quantify the effect of geometric indices derived from this algorithm on the overall classification accuracy of AAA based on whether they were electively or emergently repaired. Such quantification was performed based on a retrospective review of existing medical records of 150 AAA patients (75 electively repaired and 75 emergently repaired). Using an algorithm implemented within the MATLAB computing environment, 10 diameter- and wall thickness-related indices had a significant difference in their means when calculated relative to the AAA centerline compared to calculating them relative to the medial axis. Of these 10 indices, nine were wall thickness-related while the remaining one was the maximum diameter (Dmax). Dmax calculated with respect to the medial axis is over-estimated for both electively and emergently repaired AAA compared to its counterpart with respect to the centerline. C5.0 decision trees, a machine learning classification algorithm implemented in the R environment, were used to construct a statistical classifier. The decision trees were built by splitting the data into 70% for training and 30% for testing, and the properties of the classifier were estimated based on 1000 random combinations of the 70/30 data split. The ensuing model had average and maximum classification accuracies of 81.0 and 95.6%, respectively, and revealed that the three most significant indices in classifying AAA are, in order of importance: AAA centerline length, L2-norm of the Gaussian curvature, and AAA wall surface area. Therefore, we infer that the aforementioned three geometric indices could be used in a clinical setting to assess the risk of AAA rupture by means of a decision tree classifier. This work provides support for calculating cross-sectional diameters and wall thicknesses relative to the AAA centerline and using size and surface curvature based indices in classification studies of AAA.

Dataset Information

A Comparative Classification Analysis of Abdominal Aortic Aneurysms by Machine Learning Algorithms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets