AI identifies “rounding of the heart” as a new signal to detect cardiovascular conditions

The discovery is thanks to deep learning, one of the artificial intelligence techniques, applied to medical images of different hearts.

Cardiovascular diseases are the leading cause of death in the world, ahead of cancer, accidents and infectious diseases. Today, doctors measure heart chamber size and systolic function (the contraction of the heart during a beat) to diagnose and treat a variety of heart conditions.

A Article published in Med magazine suggests that another measure, “heart sphericity” or how round the heart is, may be useful for diagnosis.

According to the researchers, the roundness of the heart is not necessarily the problem itself, but rather an indicator. People with rounder hearts may have underlying cardiomyopathy or underlying dysfunction with heart muscle functions.

This study used big data and machine learning to explore whether other anatomical changes in the heart can improve understanding of cardiovascular risk and pathophysiology. The researchers decided to focus on sphericity because clinical experience suggests that it is associated with heart problems. Previous research has primarily focused on sphericity after the onset of heart disease and has hypothesized that sphericity may increase even before the onset of clinical heart disease.

This research used data from the UK Biobank, which includes genetic and clinical information from 500,000 people. As part of this study, MRI scans of the heart were performed on a subset of volunteers. The Californian team used data from a subset of around 38,000 UK Biobank study participants who underwent MRI scans that were considered normal at the time of the scans. The volunteers’ subsequent medical records indicated which of them subsequently developed illnesses such as cardiomyopathy, atrial fibrillation or heart failure and which did not.

Read Also:  F1 Hungarian Grand Prix Preview: Weather Forecast, Track Conditions, Tire Choices and Latest Analysis

Next, the researchers used deep learning techniques to automate the sphericity measurement. The increased sphericity of the heart appeared to be related to future heart problems. The researchers also looked at genetic factors for cardiac sphericity and found that they were consistent with genetic factors for cardiomyopathy. Using Mendelian randomization, they were able to deduce that intrinsic heart muscle disease — that is, defects not caused by heart attacks — caused the heart to be spherical.

The researchers emphasize that much more research is needed before the results of this study can be translated into clinical practice. For one thing, the connection remains speculative and would need to be confirmed with additional data. If the relationship were to be confirmed, a threshold would have to be established indicating what degree of sphericity would suggest the need for clinical interventions. The team is sharing all the data from this work and making it available to other researchers to start answering some of these questions.


Deep learning-enabled medical image analysis identifies cardiac sphericity as an early marker of cardiomyopathy and related outcomes

Recent Articles

Related News

Leave A Reply

Please enter your comment!
Please enter your name here