Artificial intelligence has developed enough that now Google is enabling its AI for predicting the death of patients even before the doctors and that too more accurately than them.
As per the new reports published in Nature, Google’s AI could soon analyse a person’s medical history and could assist doctors in making more accurate predictions relating a patient’s health and could provide related estimates on when the patient might die.
Bloomberg as an example reported that in one situation a woman reached a hospital with cancer and had fluid filled lungs. Doctors calculated a 9.3 percent chance of her dying during her hospital stay, where as Google’s AI predicted her death risk to be 19.9 percent. The woman died after a few days.
Google can make these predictions by making use of the neural networks that are good at collecting data and then using it to comprehend and enhance analysis. Besides predicting the time of death, Google’s AI technology could also predict the duration of a patient’s stay in hospitals and their chances of re-admission.
According to Daily Mail, Google’s algorithm uses patients’ data which includes their age, ethnicity, gender along side the hospital information like handwritten notes, comments, earlier diagnosis, lab results, notes in PDF, scribbles on old charts and current vital signs and make predictions accordingly.
This data gives an advantage to the hospitals for discovering new techniques for making patient care a priority, adjust their terminal plans, free healthcare workers and to catch medical emergencies well before time. These abilities assist doctors in making diagnosis.
The report also indicates that this system is much quicker and accurate that other methods used for evaluating a patient’s medical history. The AI of Google is ninety-five percent accurate, which is ten percent better than the traditional models.
The system is still in initial stages; however, Google expects that the system could one day predict death much before it occurs.