The study conducted by researchers from the Universities of Surrey, Florence, and Warwick explored how emerging technology can be utilized to help improve existing methods of detecting congestive heart failure, and have shown how AI can accurately identify CHF by quickly analyzing one ECG heartbeat.
Congestive heart failure is a chronic progressive condition that affects how blood is pumped around the body that is estimated to affect around 5 million people in America alone. According to the team of researchers clinical practitioners and health systems “urgently require efficient detection processes” as a result of “high prevalence, significant mortality rates and sustained healthcare costs”.
They team of researchers suggest that these concerns can be addressed with the use of convolutional neural networks which can be more effective for identifying patterns and structures in data; their approach combines advanced signal processing and deep machine learning tools on raw ECG signals to improve detection rates, and in another part of the study a specific CNN model was used to improve accuracy of detection while taking into account other comparable models.
Dr Sebastiano Massaro, associate professor at the University of Surrey, said: “First, by assessing ECG directly, we confirm that with AI it is possible to accurately detect CHF looking beyond heart rate variability analysis. Thus, we have in general results that are more adherent to the real behavior of the affected heart.”
“We focus on the detection of the pathology from one single heartbeat in excerpts of 5-minutes rather than in 24-hours recordings,” said Massaro. “This aspect offers a valuable potential for prospects of rapid interventions; nonetheless it is also important to keep in mind that we are talking about severe CHF patients only at the moment.”
“The application of organizational neuroscience, and specifically of neural network approaches to healthcare issues promises to open breakthrough frontiers for both clinical research and practice,” said Massaro.
