(CTN News) – According to a recent study led by an Indian-origin researcher, five types of heart failure subtypes have been identified using artificial intelligence (AI) tools that may predict future risks for individual patients based on their current risk factors.
It is a general term that refers to a condition in which the heart does not have the ability to pump blood properly around the body.
Currently, there is no accurate way of predicting the progression of heart failure by using the classification methods currently used to classify it.
In this study, published in Lancet Digital Health, researchers from University College London examined detailed anonymized patient data from over 300,000 people aged 30 or older who had been diagnosed with heart failure in the UK over a period of 20 years for their study.
Five subtypes of atrial fibrillation were identified by using several machine learning methods: onset early, late, atrial fibrillation-related (atrial fibrillation is a condition causing an irregular heart rhythm), metabolic (linked to obesity but with a low risk of cardiovascular disease), and cardiometabolic (linked to obesity and cardiovascular disease).
It was found that the patient’s risk of dying in the first year after diagnosis differed according to the subtypes of the disease.
In terms of all-cause mortality risks at one year, those were as follows: early onset (20 percent), late onset (46 percent), atrial fibrillation-related (61 percent), metabolic (11 percent), and cardiometabolic (37 percent).
Moreover, the team also developed an application that clinicians may be able to use to determine a person’s subtype of heart failure using an app developed by the group.
A better prediction of future risk and a more informed discussion with the patient may result from this process.
Our aim was to improve the classification of heart failure, so that we could better understand its likely course and communicate it to patients.
The present state of the art is that it is difficult to predict the disease’s progression for individual patients.
Professor Amitava Banerjee from the Institute of Health Informatics at University College London, said the lead author of the study. Some patients will remain stable for many years, while others will deteriorate rapidly.
In this new study, we identified five robust subtypes using multiple machine learning methods using multiple datasets in order to identify five distinct types of heart failure, which could lead to better targeted treatments as well as a different approach to potential therapies.
A key next step, according to Banerjee, would be to determine if this method of classifying heart failure can be used in a practical way to improve the quality of life of patients who suffer from heart failure.
It is important to determine if it improves risk predictions, enhances the quality of information provided by clinicians, and affects the treatment provided to patients as a result.
The app we have designed would need to be evaluated in a clinical trial or further research in order to determine whether it could be of use in routine care.
The app we have designed does need to be evaluated in a clinical trial or further research.