Additional value of biomechanical indices based on CTa for rupture risk assessment of abdominal aortic aneurysms

AI-powered assessment of biomarkers for growth prediction of abdominal aortic aneurysms

Objective

The purpose of this study was to employ biomechanics-based biomarkers to locally characterize abdominal aortic aneurysm (AAA) tissue and investigate their relation to local aortic growth by means of an artificial intelligence model.

Methods

The study focused on a population of 36 patients with AAAs undergoing serial monitoring with electrocardiogram-gated multiphase computed tomography angiography acquisitions. The geometries of the aortic lumen and wall were reconstructed from the baseline scans and used for the baseline assessment of regional aortic weakness with three functional biomarkers, time-averaged wall-shear stress, in vivo principal strain, and intra-luminal thrombus thickness. The biomarkers were encoded as regional averages on axial and circumferential sections perpendicularly to the aortic centerline. Local diametric growth was obtained as difference in diameter between baseline and follow-up at the level of each axial section. An artificial intelligence model was developed to predict accelerated aneurysmal growth with the Extra Trees algorithm used as a binary classifier where the positive class represented regions that grew more than 2.5 mm/year. Additional clinical biomarkers, such as maximum aortic diameter at baseline, were also investigated as predictors of growth.

Results

The area under the curve for the constructed receiver operating characteristic curve for the Extra Trees classifier showed a very good performance in predicting relevant aortic growth (area under the curve = 0.92), with the three biomechanics-based functional biomarkers being objectively selected as the main predictors of growth.

Conclusions

The use of features based on the functional and local characterization of the aortic tissue resulted in a superior performance in terms of growth prediction when compared with models based on geometrical assessments. With rapid growth linked to increasing risk for patients with AAAs, the ability to access functional information related to tissue weakening and disease progression at baseline has the potential to support early clinical decisions and improve disease management.

A quarter of a century biomechanical rupture risk assessment of abdominal aortic aneurysms. Achievements, clinical relevance, and ongoing developments

Abdominal aortic aneurysm (AAA) disease, the local enlargement of the infrarenal aorta, is a serious condition that causes many deaths, especially in men exceeding 65 years of age. Over the past quarter of a century, computational biomechanical models have been developed towards the assessment of AAA risk of rupture, technology that is now on the verge of being integrated within the clinical decision-making process. The modeling of AAA requires a holistic understanding of the clinical problem, in order to set appropriate modeling assumptions and to draw sound conclusions from the simulation results. In this article we summarize and critically discuss the proposed modeling approaches and report the outcome of clinical validation studies for a number of biomechanics-based rupture risk indices. Whilst most of the aspects concerning computational mechanics have already been settled, it is the exploration of the failure properties of the AAA wall and the acquisition of robust input data for simulations that has the greatest potential for the further improvement of this technology.

Prediction of Abdominal Aortic Aneurysm Growth Using Geometric Assessment of Computerized Tomography Images Acquired During the Aneurysm Surveillance Period

OBJECTIVE

We investigated the utility of geometric features for future AAA growth prediction.

BACKGROUND

Novel methods for growth prediction of AAA are recognized as a research priority. Geometric feature have been used to predict cerebral aneurysm rupture, but not examined as predictor of AAA growth.

METHODS

Computerized tomography (CT) scans from patients with infra-renal AAAs were analyzed. Aortic volumes were segmented using an automated pipeline to extract AAA diameter (APD), undulation index (UI), and radius of curvature (RC). Using a prospectively recruited cohort, we first examined the relation between these geometric measurements to patients' demographic features (n = 102). A separate 192 AAA patients with serial CT scans during AAA surveillance were identified from an ongoing clinical database. Multinomial logistic and multiple linear regression models were trained and optimized to predict future AAA growth in these patients.

RESULTS

There was no correlation between the geometric measurements and patients' demographic features. APD (Spearman r = 0.25, P < 0.05), UI (Spearman r = 0.38, P < 0.001) and RC (Spearman r =-0.53, P < 0.001) significantly correlated with annual AAA growth. Using APD, UI, and RC as 3 input variables, the area under receiver operating characteristics curve for predicting slow growth (<2.5 mm/yr) or fast growth (>5 mm/yr) at 12 months are 0.80 and 0.79, respectively. The prediction or growth rate is within 2 mm error in 87% of cases.

CONCLUSIONS

Geometric features of an AAA can predict its future growth. This method can be applied to routine clinical CT scans acquired from patients during their AAA surveillance pathway.

Geometric and biomechanical modeling aided by machine learning improves the prediction of growth and rupture of small abdominal aortic aneurysms

It remains difficult to predict when which patients with abdominal aortic aneurysm (AAA) will require surgery. The aim was to study the accuracy of geometric and biomechanical analysis of small AAAs to predict reaching the threshold for surgery, diameter growth rate and rupture or symptomatic aneurysm. 189 patients with AAAs of diameters 40–50 mm were included, 161 had undergone two CTAs. Geometric and biomechanical variables were used in prediction modelling. Classifications were evaluated with area under receiver operating characteristic curve (AUC) and regressions with correlation between observed and predicted growth rates. Compared with the baseline clinical diameter, geometric-biomechanical analysis improved prediction of reaching surgical threshold within four years (AUC 0.80 vs 0.85, p = 0.031) and prediction of diameter growth rate (r = 0.17 vs r = 0.38, p = 0.0031), mainly due to the addition of semiautomatic diameter measurements. There was a trend towards increased precision of volume growth rate prediction (r = 0.37 vs r = 0.45, p = 0.081). Lumen diameter and biomechanical indices were the only variables that could predict future rupture or symptomatic AAA (AUCs 0.65–0.67). Enhanced precision of diameter measurements improves the prediction of reaching the surgical threshold and diameter growth rate, while lumen diameter and biomechanical analysis predicts rupture or symptomatic AAA.

Comparison of peak wall stress and peak wall rupture index in ruptured and asymptomatic intact abdominal aortic aneurysms

BACKGROUND

Previous studies have suggested that finite element analysis (FEA) can estimate the rupture risk of an abdominal aortic aneurysm (AAA); however, the value of biomechanical estimates over measurement of AAA diameter alone remains unclear. This study aimed to compare peak wall stress (PWS) and peak wall rupture index (PWRI) in participants with ruptured and asymptomatic intact AAAs.

METHODS

The reproducibility of semiautomated methods for estimating aortic PWS and PWRI from CT images was assessed. PWS and PWRI were estimated in people with ruptured AAAs and those with asymptomatic intact AAAs matched by orthogonal diameter on a 1 : 2 basis. Spearman's correlation coefficient was used to assess the association between PWS or PWRI and AAA diameter. Independent associations between PWS or PWRI and AAA rupture were identified by means of logistic regression analyses.

RESULTS

Twenty individuals were included in the analysis of reproducibility. The main analysis included 50 patients with an intact AAA and 25 with a ruptured AAA. Median orthogonal diameter was similar in ruptured and intact AAAs (82·3 (i.q.r. 73·5-92·0) versus 81·0 (73·2-92·4) mm respectively; P = 0·906). Median PWS values were 286·8 (220·2-329·6) and 245·8 (215·2-302·3) kPa respectively (P = 0·192). There was no significant difference in PWRI between the two groups (P = 0·982). PWS and PWRI correlated positively with orthogonal diameter (both P < 0·001). Participants with high PWS, but not PWRI, were more likely to have a ruptured AAA after adjusting for potential confounders (odds ratio 5·84, 95 per cent c.i. 1·22 to 27·95; P = 0·027). This association was not maintained in all sensitivity analyses.

CONCLUSION

High aortic PWS had an inconsistent association with greater odds of aneurysm rupture in patients with a large AAA.

Systematic Review and Meta-Analysis of Peak Wall Stress and Peak Wall Rupture Index in Ruptured and Asymptomatic Intact Abdominal Aortic Aneurysms

Background Prior studies have suggested aortic peak wall stress (PWS) and peak wall rupture index (PWRI) can estimate the rupture risk of an abdominal aortic aneurysm (AAA), but whether these measurements have independent predictive ability over assessing AAA diameter alone is unclear. The aim of this systematic review was to compare PWS and PWRI in participants with ruptured and asymptomatic intact AAAs of similar diameter. Methods and Results Web of Science, Scopus, Medline, and The Cochrane Library were systematically searched to identify studies assessing PWS and PWRI in ruptured and asymptomatic intact AAAs of similar diameter. Random-effects meta-analyses were performed using inverse variance-weighted methods. Leave-one-out sensitivity analyses were conducted to assess the robustness of findings. Risk of bias was assessed using a modification of the Newcastle-Ottawa scale and standard quality assessment criteria for evaluating primary research papers. Seven case-control studies involving 309 participants were included. Meta-analyses suggested that PWRI (standardized mean difference, 0.42; 95% CI, 0.14-0.70; P=0.004) but not PWS (standardized mean difference, 0.13; 95% CI, -0.18 to 0.44; P=0.418) was greater in ruptured than intact AAAs. Sensitivity analyses suggested that the findings were not dependent on the inclusion of any single study. The included studies were assessed to have a medium to high risk of bias. Conclusions Based on limited evidence, this study suggested that PWRI, but not PWS, is greater in ruptured than asymptomatic intact AAAs of similar maximum aortic diameter.

Matched Cohort Study of Cataract Surgery With and Without Trabecular Microbypass Stent Implantation in Primary Angle-Closure Glaucoma

PURPOSE

To compare 1-year outcomes of phacoemulsification alone (phaco-only) vs phacoemulsification with implantation of 2 trabecular microbypass stents (iStent or iStent inject; phaco-stent) in eyes with primary angle-closure glaucoma (PACG).

DESIGN

Retrospective matched clinical cohort study.

METHODS

PACG eyes that underwent phaco-only vs phaco-stent at a single ophthalmology center. Groups were matched for baseline intraocular pressure (IOP) and medication use with a tolerance of ±2 mm Hg and ±1 medication, respectively. Primary outcomes included postoperative change in the mean IOP and medications. One-year outcomes were assessed using generalized estimating equations corrected for baseline intergroup differences.

RESULTS

One hundred fifty-eight eyes (79 per group) were included. At 1 year, IOP decreased by 13% (from 16.8 ± 3.1 mm Hg preoperatively) in the phaco-only group (P < .001) and by 27% (from 17.6 ± 3.2 mm Hg) in the phaco-stent group (P < .001). Medication use decreased by 11% (from 1.8 ± 1.3 medications preoperatively) in the phaco-only group (P < .001) and by 46% (from 2.2 ± 1.2 medications) in the phaco-stent group (P < .001). The phaco-stent group experienced significantly larger reductions in IOP and medications compared with the phaco-only group (P < .001). The incidence of IOP spikes was significantly greater in the phaco-only group (18%) compared with the phaco-stent group (4%; P = .005). Safety was favorable with few transient postoperative adverse events.

CONCLUSION

The results of this study highlight that phacoemulsification with implantation of 2 trabecular microbypass stents is more effective and possibly more protective than phaco-only in PACG eyes, as evidenced by significantly larger IOP and medication reductions and smaller incidences of IOP spikes among the phaco-stent eyes.

Predictors of Abdominal Aortic Aneurysm Risks

Computational biomechanics via finite element analysis (FEA) has long promised a means of assessing patient-specific abdominal aortic aneurysm (AAA) rupture risk with greater efficacy than current clinically used size-based criteria. The pursuit stems from the notion that AAA rupture occurs when wall stress exceeds wall strength. Quantification of peak (maximum) wall stress (PWS) has been at the cornerstone of this research, with numerous studies having demonstrated that PWS better differentiates ruptured AAAs from non-ruptured AAAs. In contrast to wall stress models, which have become progressively more sophisticated, there has been relatively little progress in estimating patient-specific wall strength. This is because wall strength cannot be inferred non-invasively, and measurements from excised patient tissues show a large spectrum of wall strength values. In this review, we highlight studies that investigated the relationship between biomechanics and AAA rupture risk. We conclude that combining wall stress and wall strength approximations should provide better estimations of AAA rupture risk. However, before personalized biomechanical AAA risk assessment can become a reality, better methods for estimating patient-specific wall properties or surrogate markers of aortic wall degradation are needed. Artificial intelligence methods can be key in stratifying patients, leading to personalized AAA risk assessment.

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

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.