Google AI now predicts when a patient will die

A women at her last stage of the breast cancer came to a hospital with fluids flooding her lungs. She late on got a radiology scan after which the hospital’s computers finds her chance for staying live with an estimate of 9.3% but Google’s new algorithm modified it as 19.9%. she died within a matter of days.

The incident was later on published by Google in May thus educating people about the importance of health care potential of neural network which is a form of artificial intelligence software. This new network will forecast whether a patient may stay in hospitals, their re-admission and chances of their death.

Google said in a study about the Neural network that, “These AI based predictive models outperformed traditional, clinically used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios”.

The basic difference between current warning systems and Google AI is that the current warning system can read the data present in only electronic health records (EHR) while Google AI can read all the data on EHR as well as scribbled notes, huge amount of structured and semi-structured data and notes buried in PDFs.

With the new invention, Google will now step into the healthcare market too. Despite being the first inventor of the commercially viable AI technology as well as the king of the internet services, the medical brain team’s research will break into the new brand market.

Related posts

Leave a Comment