Registry aims to help predict patients' risk for hypertrophic cardiomyopathy
Researchers have detailed the first results from the world's largest comprehensive study of hypertrophic cardiomyopathy (HCM), an abnormal thickening of the heart that often goes undiagnosed. Investigators developed and examined a registry of more than 2,750 patients with the condition at 44 sites in six countries. The team combined imaging, genetic analysis and biomarker data with traditional clinical information for this analysis. Most prior HCM analyses have been retrospective, but with the Hypertrophic Cardiomyopathy Registry, researchers have collected a range of data from the study participants to gain a larger perspective. The researchers believe that the registry ultimately will let them identify clues to determine which patients are at greatest risk and which treatments will benefit different groups of patients the most. Of note, early findings suggest that patients largely can be grouped into two buckets. Those with a clearly defined genetic mutation tended to have more scarring of the heart muscle, while patients without such a mutation tended to have no scar and more obstruction of blood flow. That information potentially could help researchers predict patients' risk of sudden cardiac death and heart failure and determine the best treatment strategies. "It really changes the way we think about patients. We can categorize them more easily," said Christopher Kramer, a cardiologist at University of Virginia Health and a co-principal investigator of the study. "The more we can understand and group patients into categories, the better we will be able to learn what the best therapies are." The article "Distinct Subgroups in Hypertrophic Cardiomyopathy in the NHLBI HCM Registry" appears in the Nov. 12, 2019, issue of the Journal of the American College of Cardiology.
AI can examine ECGs to forecast irregular heartbeat, death risk
Artificial intelligence (AI) can examine electrocardiogram (ECG) test results to identify patients who are at higher risk of developing arrhythmia or of dying within the next year, according to two preliminary studies to be presented at the American Heart Association's Scientific Sessions 2019, taking place Nov. 16 through 18 in Philadelphia. Researchers used more than 2 million ECG results from roughly three decades of archived medical records in Pennsylvania/New Jersey's Geisinger Health System to train deep neural networks. Both studies are among the first to use AI to predict future events from an ECG rather than to detect current health problems. "This is exciting and provides more evidence that we are on the verge of a revolution in medicine where computers will be working alongside physicians to improve patient care," said Brandon Fornwalt, senior author on both studies and associate professor and chair of the department of imaging science and innovation at Geisinger in Danville, Pa. According to one poster presentation, researchers speculated that a deep learning model could predict atrial fibrillation before it develops. When looking at 1.1 million ECGs that did not indicate the presence of atrial fibrillation in more than 237,000 patients, researchers used computational hardware to train a deep neural network to analyze 15 segments of data for each ECG. They determined that within the top 1% of high-risk patients, as predicted by the neural network, one of every three people was diagnosed with atrial fibrillation within a year. "Currently, there are limited methods to identify which patients will develop AF within the next year, which is why, many times, the first sign of AF is a stroke," said senior author Christopher Haggerty, assistant professor in the Department of Imaging Science and Innovation at Geisinger. "We hope that this model can be used to identify patients with atrial fibrillation very early so they can be treated to prevent stroke." A second presentation looked to help identify patients most likely to die of any cause within a year, Geisinger researchers analyzed the results of 1.77 million ECGs and other records from almost 400,000 patients. They used this data to compare machine learning-based models that either directly analyzed the raw ECG signals or relied on aggregated human-derived measures and commonly diagnosed disease patterns. The neural network model that directly analyzed the ECG signals was found to be superior for predicting one-year risk of death.
Be sure to get your Zzzs
There are plenty of people not getting adequate sleep. Analyzing data collected from wearable trackers, researchers from the SingHealth Duke-NUS Institute of Precision Medicine (PRISM) and the National Heart Centre Singapore (NHCS) recently demonstrated that chronic sleep deprivation is associated with increased cardiovascular disease risk markers and accelerated biological aging. The PRISM-NHCS team analyzed the sleep patterns of Singaporeans through data collected from wearable technology, with more than 480 healthy volunteers using Fitbit trackers and submitting one week's sleep data for the study. To estimate biological age, the team analyzed the volunteers' whole-genome data to estimate their telomere lengths. The team found that the 7% of volunteers who slept less than five hours a night were twice as likely to have shortened telomeres vs. those who exceeded the recommended sleep amount of seven hours. They also had increased cardiovascular risk factors, such as higher body mass indexes and waist circumferences. "Consumer wearables have the capacity to capture a lot of data from individuals in their day-to-day life without being intrusive," said senior author Patrick Tan, director, SingHealth Duke-NUS PRISM and Professor, Cancer and Stem Cell Biology Program, Duke-NUS Medical School, on the use of consumer-grade wearable technology for research. The findings have been published in the journal Communications Biology. "Digital phenotyping by consumer wearables identifies sleep-associated markers of cardiovascular disease risk and biological aging," appeared Oct. 4, 2019.