Publications by authors named "S Lubitz"
The 12-lead electrocardiogram (ECG) is inexpensive and widely available. Whether conditions across the human disease landscape can be detected using the ECG is unclear. We developed a deep learning denoising autoencoder and systematically evaluated associations between ECG encodings and ~1,600 Phecode-based diseases in three datasets separate from model development, and meta-analyzed the results.
View Article and Find Full Text PDFTo broaden our understanding of bradyarrhythmias and conduction disease, we performed common variant genome-wide association analyses in up to 1.3 million individuals and rare variant burden testing in 460,000 individuals for sinus node dysfunction (SND), distal conduction disease (DCD) and pacemaker (PM) implantation. We identified 13, 31 and 21 common variant loci for SND, DCD and PM, respectively.
View Article and Find Full Text PDFArticle Synopsis
- Medical record review by physician committees is the current standard for identifying cardiovascular outcomes in clinical trials, but it's time-consuming and inconsistent.
- A new AI model called "HF-NLP" was developed to automatically assess heart failure outcomes, tested on data from international trials, including the DELIVER trial.
- The AI achieved 83% agreement with expert committee decisions, and when supplemented with human reviews for uncertain cases, it could reach 91% agreement while significantly reducing workload.
Article Synopsis
- TTN encodes the protein titin and is commonly associated with rare variants in patients diagnosed with atrial fibrillation (AF) during genetic testing.
- The study compared characteristics and outcomes of patients with AF having pathogenic TTN variants to those without such variants, revealing that TTN(+) patients experience more persistent AF and require more cardioversions.
- Findings indicate that nearly 50% of TTN(+) AF patients develop serious heart issues, emphasizing the importance of specialized evaluation and management for these individuals.
Article Synopsis
- The study investigates how rare non-coding genetic variations affect complex traits, specifically focusing on human height by analyzing data from over 333,100 individuals across three large datasets.
- Researchers found 29 significant rare variants linked to height, with impacts ranging from a decrease of 7 cm to an increase of 4.7 cm, after considering previously known variants.
- The team also identified specific non-coding variants near key genes associated with height, demonstrating a new method for understanding the effects of rare variants in regulatory regions using whole-genome sequencing.