Citations

Introduction: Multiple supervised machine learning (ML) models have been used to assess the outcomes in patients who underwent transcatheter aortic valve replacement (TAVR) procedure, however, supervised ML requires significant expert time for annotation and labeling. Unsupervised ML can provide insight into the dataset without this process. Hypothesis: We hypothesized that unsupervised k-means ML can identify distinct phenogroups in TAVR patients that can predict prognosis without a priori knowledge. Methods: A retrospective study of patients who underwent TAVR (Jan 2012-Dec 2017) at Mayo Clinic sites was performed. Baseline demographics, EKG, cardiac computed tomography, and echocardiographic data were abstracted. Variables with ≥ 20% missing data were excluded. Prior to the k-means clustering algorithm, principal component analysis was used for dimensionality reduction. Input variables were analyzed according to k-means determined clusters. Kaplan-Meier survival analysis was used to assess the survival of each cluster. Results: A total of 1071 consecutive patients were included in the final analysis. A total of 141 unique variables were used. The mean age was 80.7±8.6 years, 58.2% were male. Patients were divided into 2 distinct phenogroups. In the two phenogroups, cluster 0 (n=639) has significantly better survival than cluster 1 (n= 432) (log-rank p=0.0002) (Figure 1). Cluster 1 has features of low flow-low gradient (LFLG) hemodynamic profile, including lower systolic blood pressure, mean aortic valve gradient, peak aortic valve velocity, and worse LV ejection fraction. Conclusions: Unsupervised ML can identify distinct phenotype features with prognostic significance in TAVR patients. Our results suggest that patients with LFLG hemodynamic features are intrinsically different from other TAVR patients and require more attention to avoid adverse outcomes.

Introduction: Syncope was reported to be associated with sudden cardiac death in heart failure patients regardless of the etiology of syncope. We attempted to use unsupervised machine learning to identify the prognostically distinct phenotypes in a large population of heart failure patients presenting with syncope. Hypothesis: Unsupervised machine learning can identify prognostically distinct phenotypes in a large population of heart failure patients presenting with syncope. Methods: Patients with a diagnosis of systolic or diastolic heart failure and syncope at Mayo Clinic sites were identified for baseline and follow-up data. Variables with ≥ 20% missing data were excluded; a total of 57 variables were used for k-means clustering analysis. Patients were grouped according to 4 distinct k-means determined clusters for statistical analysis. Results: A total of 3,557 patients with proper follow-up and echocardiography data were included in the final analysis. The median follow-up duration was 36.3 months, and 2908 (81.7%) patients died. Cluster 2 (n= 680) was identified as the cluster with the highest mortality (92%), which contained patients with high proportions of cardiovascular comorbidities, especially chronic kidney disease. The Kaplan-Meier survival curves are shown in Figure 1. In multivariate Cox regression analysis, the presence of CKD independently predicts mortality (HR 1.62, 95%CI 1.50-1.76, p <0.0001) after adjusting for age and sex. Conclusions: Unsupervised machine learning identified patients with CKD, CAD and DM are high-risk phenotypes in heart failure patients with syncope, and can be used to guide future studies in this highly heterogeneous, unexplored population.

Introduction: Unsupervised machine learning (ML) can provide insight into a dataset without significant expert time for annotation and labeling, but was never used to assess the heterogeneous MitraClip population. Hypothesis: We hypothesized that unsupervised k-means ML could identify prognostically-distinct phenogroups in MitraClip patients without a priori knowledge of the dataset. Methods: Patients who underwent MitraClip (June 2014-September 2020) at Mayo Clinic sites were identified from the Mayo institutional NCDR database for baseline and follow-up data. Variables with ≥ 50% missing data were excluded. A k-means algorithm was used for clustering analysis. Input variables were grouped according to 4 distinct k-means determined clusters. Kaplan-Meier survival analysis was used to assess the survival of each cluster. Results: A total of 389 consecutive patients were included in the final analysis, and 95 unique variables were used. The mean age was 80.3±8.7 years; 256 (65.8%) were male. The mean STS MV Replace score was 9.1±5.7. Fifty-five (14.5%) patients died during the mean follow duration (185 days). Kaplan-Meier analysis showed significant survival differences among clusters (Figure 1). The cluster (cluster 2, n=120) with best survival performance (all-cause mortality 3.3%) had features of non-calcified mitral leaflets, nonischemic cardiomyopathy, and less than moderate tricuspid regurgitation. Cluster 3 (n=66) had the worst survival performance (all-cause mortality 27.3%) and opposite features than cluster 2. Conclusions: Unsupervised ML can identify distinct phenotype clusters with prognostic significance in MitraClip patients. Patients with non-calcified mitral leaflets, nonischemic cardiomyopathy, and less than moderate tricuspid regurgitation are intrinsically different from other MitraClip patients and should be considered the potential optimal candidates.

Introduction: Ischemic stroke (IS) causes substantial morbidity and mortality in patients undergoing percutaneous coronary intervention (PCI) with a 5 yr incidence ~ 3%. We sought the test the accuracy of Machine learning (ML) algorithms in predicting IS in patients undergoing PCI. Methods: Mayo Clinic CathPCI registry data were retrospectively analyzed from Jan 2003 - June 2018 including 21,872 patients who underwent PCI. The cohort was randomly divided into a training sample (75%, n=16404) and a unique test sample (25%, n=5468) prior to model generation. The risk prediction model was generated utilizing a random forest algorithm (RF model) on 188 unique variables to predict the risk of IS at 6-month, 1, 2, and 5-year post PCI. Conventional risk factors for stroke were used for logistic regression. The receiver operating characteristic (ROC) curve and area under the curve for the RF and logistic regression models were compared for the test cohort. Results: The mean age was 66.9 ± 12.4 years, and 71% were male. Patient demographics and outcomes are shown in Table 1 . The ROC area under the curve for the RF model was superior compared to the logistic regression model in predicting IS at 6 months, 1,2 and 5 yrs for the test cohort ( Figure 1 .) Conclusions: The RF model accurately predicts short and long term risk of IS and outperforms logistic regression analysis in patients undergoing PCI.