All of us are scared and at the same time, curious to know what the future holds for us. When will I get married? How many kids will I have? What will be my next great job? Will I be able to buy that dream house? But of all these questions, the most important one is- when will I die?
This question becomes crucial for many as it’s the last knell to the coffin, quite literally. But can we know about all these?
Can someone really answer these questions for us? While most of the future is uncertain, is it now possible to predict the approximate time of death! We all have to bow down to Artificial Intelligence for that.
After scrutinizing ECG tests, it is now possible for Artificial Intelligence to predict when a specific patient might die in the coming year.
To claim this impossible stance, researchers from the Geisinger Health System in Pennsylvania have analyzed the results of 1.77 million ECGs and other data from 400,000 patients.
The data was then used to compare results from machine-learning models that either studied the ECG samples directly or depended on aggregated human measures. The neural network model featured by the AI showed far more superior results for identifying the one-year death risk.
To recheck the AI results, three renowned cardiologists were made to look into the ECGs, which initially read normally. The cardiologists diagnosed the same risk patterns as highlighted by artificial learning.
As per many cardiologists, the importance of this study is over the tops. The study can single-handedly change the course of interpreting ECG results in the future.
Another study in this field concluded that AI-based study could interpret ECG results to predict the risks that might show up in the future. It will now be possible to detect the early signs of arrhythmia as and when they arise.