Citations

Background: Deep learning algorithms derived in homogeneous populations may be poorly generalizable and have the potential to reflect, perpetuate, and even exacerbate racial/ethnic disparities in health and health care. In this study, we aimed to (1) assess whether the performance of a deep learning algorithm designed to detect low left ventricular ejection fraction using the 12-lead ECG varies by race/ethnicity and to (2) determine whether its performance is determined by the derivation population or by racial variation in the ECG. Methods: We performed a retrospective cohort analysis that included 97 829 patients with paired ECGs and echocardiograms. We tested the model performance by race/ethnicity for convolutional neural network designed to identify patients with a left ventricular ejection fraction ≤35% from the 12-lead ECG. Results: The convolutional neural network that was previously derived in a homogeneous population (derivation cohort, n=44 959; 96.2% non-Hispanic white) demonstrated consistent performance to detect low left ventricular ejection fraction across a range of racial/ethnic subgroups in a separate testing cohort (n=52 870): non-Hispanic white (n=44 524; area under the curve [AUC], 0.931), Asian (n=557; AUC, 0.961), black/African American (n=651; AUC, 0.937), Hispanic/Latino (n=331; AUC, 0.937), and American Indian/Native Alaskan (n=223; AUC, 0.938). In secondary analyses, a separate neural network was able to discern racial subgroup category (black/African American [AUC, 0.84], and white, non-Hispanic [AUC, 0.76] in a 5-class classifier), and a network trained only in non-Hispanic whites from the original derivation cohort performed similarly well across a range of racial/ethnic subgroups in the testing cohort with an AUC of at least 0.930 in all racial/ethnic subgroups. Conclusions: Our study demonstrates that while ECG characteristics vary by race, this did not impact the ability of a convolutional neural network to predict low left ventricular ejection fraction from the ECG. We recommend reporting of performance among diverse ethnic, racial, age, and sex groups for all new artificial intelligence tools to ensure responsible use of artificial intelligence in medicine.

Atrial fibrillation (AF) is an important comorbidity in patients with end-stage renal disease (ESRD) undergoing dialysis that portends increased health care utilization, morbidity, and mortality in this already high-risk population. Patients with ESRD have a particularly high stroke risk, which is further compounded by AF. However, the role of anticoagulation for stroke prophylaxis in ESRD and AF is debated. The ESRD population presents a unique challenge because of the combination of elevated stroke and bleeding risks. Warfarin has been traditionally used in this population, but it is associated with significant risks of minor and major bleeding, particularly intracranial, thus leading many clinicians to forgo anticoagulation altogether. When anticoagulation is prescribed, rates of adherence and persistence are poor, leaving many patients untreated. The direct oral anticoagulants (DOACs) may offer an alternative to warfarin in ESRD patients, but these agents have not been extensively studied in this population and uncertainties regarding comparative effectiveness (versus warfarin, each other, and no treatment) remain. In this review, we discuss the current evidence on the risk and benefits of anticoagulants in this challenging population and comparisons between warfarin and DOACs, and review future directions including options for non-pharmacologic stroke prevention.

AbstractBackgroundThe accurate separation of undifferentiated wide complex tachycardias (WCTs) into ventricular tachycardia (VT) or supraventricular wide complex tachycardia (SWCT) using conventional, manually‐applied 12‐lead electrocardiogram (ECG) interpretation methods is difficult.PurposeWe sought to devise a new WCT differentiation method that operates solely on automated measurements routinely provided by computerized ECG interpretation software.MethodsIn a two‐part analysis, we developed and validated a logistic regression model (ie, VT Prediction Model) that utilizes routinely available computerized measurements derived from patients' paired WCT and baseline ECGs.ResultsThe derivation cohort consisted of 601 paired WCT (273 VT, 328 SWCT) and baseline ECGs from 421 patients. The VT Prediction Model, composed of WCT QRS duration (ms) (P < .0001), QRS duration change (ms) (P < .0001), QRS axis change (°) (P < .0001) and T axis change (°) (P < .0001), yielded effective VT and SWCT differentiation (area under the curve [AUC]: 0.924; confidence interval [CI]: 0.903‐0.944) for the derivation cohort. The validation cohort comprised 241 paired WCT (97 VT, 144 SWCT) and baseline ECGs from 177 patients. The VT Prediction Model's implementation on the validation cohort yielded effective WCT differentiation (AUC: 0.900; CI: 0.862‐0.939) with overall accuracy, sensitivity, and specificity of 85.0%, 80.4%, and 88.2%, respectively.ConclusionThe VT Prediction Model is an example of how readily available ECG measurements may be used to distinguish VT and SWCT effectively. Further study is needed to develop and refine newer WCT differentiation approaches that utilize computerized measurements provided by ECG interpretation software.