Summary:
Heart failure is a serious condition where the heart struggles to pump enough blood to meet the body’s needs, leading to symptoms such as fatigue, weakness, and swelling in the legs and feet, and ultimately, it can be fatal. It is a progressive disease, underscoring the importance of healthcare providers identifying patients at high risk for worsening outcomes. Recently, researchers have introduced a powerful new risk assessment tool designed to predict the prognosis of heart failure patients.
Developed by researchers at the University of Virginia Health System, this innovative model leverages machine learning (ML) and artificial intelligence (AI) to evaluate individual risks of developing serious complications associated with heart failure. Drawing from anonymized data of thousands of patients from previous heart failure clinical trials sponsored by the National Heart, Lung, and Blood Institute, this tool has proven more effective than existing predictors in forecasting outcomes across a wide spectrum of patients. It assesses probabilities such as the need for cardiac surgery or transplantation, risk of rehospitalization, and mortality risk.
The success of the model lies in its integration of ML/AI technologies and inclusion of hemodynamic clinical data, detailing how blood moves through the heart, lungs, and other parts of the body. Researchers emphasize that applying this model allows physicians to tailor treatment more precisely to each patient’s needs, potentially extending and enhancing their quality of life.
«Heart failure is a progressive condition that impacts not just quality but also quantity of life. Each patient falls along a risk spectrum for adverse outcomes,» noted heart failure expert Dr. Sula Mazimba. «Identifying each patient’s risk profile promises to assist physicians in optimizing therapies to improve outcomes.»
«This model represents a significant advancement as it processes complex data sets and can make decisions even amidst missing or conflicting factors,» added researcher Josephine Lamp from the University of Virginia’s Department of Computer Science. «It’s truly exciting because the model intelligently presents and summarizes risk factors, thereby reducing decision burdens and enabling quicker treatment decisions for physicians.»
Article written by Diego Domingo| Image by Leuko
17/06/2024
Source:
ConSalud
https://www.consalud.es/saludigital/tecnologia-sanitaria/detecta-infecciones-pacientes-cancer-sin-hacer-analisis-sangre_145253_102.html