This Deep Learning Network can reduce heart attacks through early prediction

Healthy heart is the key to a healthy and long life. Modern lifestyles have made human beings a victim of cardiovascular diseases. The tendency to smoke often, eat unhealthy, and not following a proper exercise schedule is making human heart failure prone to failure.

 

Some cardiovascular diseases like heart attack and heart failure can gulp down a human's life within a few seconds. Researchers have developed a deep learning network that can predict a patient’s risk of encountering a cardiac arrest in future.

 

This deep learning network can predict extreme cardiac events like heart attacks. The research was conducted on 20,000 patients. During the testing period the deep learning network was capable of predicting major adverse cardiac events (MACE) using a single photon emission tomography (SPECT) and myocardial perfusion imaging (MFI).

 

The deep learning method also identified polar image regions. Identification of polar image regions will help physicians to visually interpret the patients who are at the risk of cardiac arrest. The network gave each patient a score to predict the chances of heart failure in the future. Patients were then tracked for a period of 4-7 years.

 

The network was developed using polar image inputs of raw perfusion and gated derived maps of motion thickening face angle, and amplitude. These factors were combinesd with age, sex and end systolic volumes to get the accurate prediction.

 

(Must Check: 4 Uses of AI In Cardiology)

 

The method traced and highlighted specific regions of the heart that were associated with MACE risk. Patients who got a 9.7 percentile MACE rate were at extreme risk. Deep learning networks simplified the process of MACE prediction. The polar map images provide accurate results. The network was more accurate compared to its former networks. 

 

As reported by Health IT Analytics Ananya Singh, MS, a research software engineer in the Slomka Lab at Cedars-Sinai Medical Center in Los Angeles, CA, said, “These findings show that artificial intelligence could be incorporated in standard clinical workstations to assist physicians in accurate and fast risk assessment of patients undergoing SPECT MPI scans.”

 

According to the statistics cardiovascular diseases are responsible for 17.9 billion deaths every year. The new method can be used to make the patients aware about advancing risk. Deep Learning has transformed the healthcare sector. It has lent a helping hand to the healthcare professionals worldwide, it can also predict the advancing risk of brain related disorders.