Featured Publications

Background To describe outcomes of high-risk patients with coronavirus disease 2019 (COVID-19) treated with sotrovimab, other monoclonal antibodies (mAbs), or antivirals, and patients who did not receive early COVID-19 treatment. We also evaluate the comparative effectiveness of sotrovimab versus no treatment in preventing severe clinical outcomes. Methods This observational retrospective cohort study analyzed Mayo Clinic electronic health records. Non-hospitalized adult patients diagnosed with COVID-19 from May 26, 2021 and April 23, 2022 and at high risk of COVID-19 progression were eligible. The primary outcome was 29-day all-cause hospitalization and/or death. Outcomes were described for patients treated with sotrovimab, other mAbs, or antivirals, and eligible but untreated patients, and compared between sotrovimab-treated and propensity score (PS)-matched untreated cohorts. Results We included 35,485 patients (sotrovimab, 1369; other mAbs, 6488; antivirals, 133; high-risk untreated, 27,495). A low proportion of patients treated with sotrovimab (n = 33/1369, 2.4%), other mAbs (n = 147/6488, 2.3%), or antivirals (n = 2/133, 1.5%) experienced all-cause hospitalization or death. Among high-risk untreated patients, the percentage of all-cause hospitalization or death was 3.3% (n = 910/27,495). In the PS-matched analysis, 2.5% (n = 21/854) of sotrovimab-treated patients experienced all-cause hospitalization and/or death versus 2.8% (n = 48/1708) of untreated patients (difference, –0.4%; p = 0.66). Significantly fewer sotrovimab-treated patients required intensive care unit admission (0.5% vs 1.8%; difference, –1.3%; p = 0.002) or respiratory support (3.5% vs 8.7%; difference, –5.2%; p < 0.001). Conclusions There was no significant difference in the proportion of sotrovimab-treated and PS-matched untreated patients experiencing 29-day all-cause hospitalization or mortality, although significantly fewer sotrovimab-treated patients required intensive care unit admission or respiratory support.

Background Early diagnosis of pulmonary hypertension (PH) is critical for effective treatment and management. We aimed to develop and externally validate an artificial intelligence algorithm that could serve as a PH screening tool, based on analysis of a standard 12-lead ECG. Methods The PH Early Detection Algorithm (PH-EDA) is a convolutional neural network developed using retrospective ECG voltage–time data, with patients classified as “PH-likely” or “PH-unlikely” (controls) based on right heart catheterisation or echocardiography. In total, 39 823 PH-likely patients and 219 404 control patients from Mayo Clinic were randomly split into training (48%), validation (12%) and test (40%) sets. ECGs taken within 1 month of PH diagnosis (diagnostic dataset) were used to train the PH-EDA at Mayo Clinic. Performance was tested on diagnostic ECGs within the test sets from Mayo Clinic (n=16 175/87 998 PH-likely/controls) and Vanderbilt University Medical Center (VUMC; n=6045/24 256 PH-likely/controls). In addition, performance was tested on ECGs taken 6–18 months (pre-emptive dataset), and up to 5 years prior to a PH diagnosis at both sites. Results Performance testing yielded an area under the receiver operating characteristic curve (AUC) of 0.92 and 0.88 in the diagnostic test sets at Mayo Clinic and VUMC, respectively, and 0.86 and 0.81, respectively, in the pre-emptive test sets. The AUC remained a minimum of 0.79 at Mayo Clinic and 0.73 at VUMC up to 5 years before diagnosis. Conclusion The PH-EDA can detect PH at diagnosis and 6–18 months prior, demonstrating the potential to accelerate diagnosis and management of this debilitating disease.

Electronic Health Records (EHRs) provide a rich source of medical information across different modalities such as electrocardiograms (ECG), structured EHRs (sEHR), and unstructured EHRs (text). Inspired by the fact that many cardiac and non-cardiac diseases influence the behavior of the ECG, we leverage structured EHRs and unstructured EHRs from multiple sources by pairing with ECGs and propose a set of three new multi-modal contrastive learning models that combine ECG, sEHR, and text modalities. The performance of these models is compared against different baseline models such as supervised learning models trained from scratch with random weights initialization, and self-supervised learning models trained only on ECGs. We pre-train the models on a large proprietary dataset of about 9 million ECGs from around 2.4 million patients and evaluate the pre-trained models on various downstream tasks such as classification, zero-shot retrieval, and out-of-distribution detection involving the prediction of various heart conditions using ECG waveforms as input, and demonstrate that the models presented in this work show significant improvements compared to all baseline modes.

The automated analysis of medical time series, such as the electrocardiogram (ECG), electroencephalogram (EEG), pulse oximetry, etc, has the potential to serve as a valuable tool for diagnostic decisions, allowing for remote monitoring of patients and more efficient use of expensive and time-consuming medical procedures. Deep neural networks (DNNs) have been demonstrated to process such signals effectively. However, previous research has primarily focused on classifying medical time series rather than attempting to regress the continuousvalued physiological parameters central to diagnosis. One significant challenge in this regard is the imbalanced nature of the dataset, as a low prevalence of abnormal conditions can lead to heavily skewed data that results in inaccurate predictions and a lack of certainty in such predictions when deployed. To address these challenges, we propose HypUC, a framework for imbalanced probabilistic regression in medical time series, making several contributions. (i) We introduce a simple kernel density-based technique to tackle the imbalanced regression problem with medical time series. (ii) Moreover, we employ a probabilistic regression framework that allows uncertainty estimation for the predicted continuous values. (iii) We also present a new approach to calibrate the predicted uncertainty further. (iv) Finally, we demonstrate a technique to use calibrated uncertainty estimates to improve the predicted continuous value and show the efficacy of the calibrated uncertainty estimates to flag unreliable predictions. HypUC is evaluated on a large, diverse, real-world dataset of ECGs collected from millions of patients, outperforming several conventional baselines on various diagnostic tasks, suggesting potential use-case for the reliable clinical deployment of deep learning models and a prospective clinical trial. Consequently, a hyperkalemia diagnosis algorithm based on HypUC is going to be the subject of a real-world clinical prospective study.