AI knows when you’re going to die. Scary, right?
A new paper published in Nature suggests that feeding electronic health record data to a deep learning model could substantially improve the accuracy of projected outcomes. In trials using data from two american hospitals, researchers were able to show that these algorithms could predict a patient’s length of stay and time of discharge, but also the time of death.
The team took a different approach to building predictive statistical models by considering a ‘representation’ of all a patient’s health records, including clinical notes, rather than removing most of a patient’s information from the analysis. Eighty percent of the effort in creating an analytic model is in cleaning the data, so it could provide a way to scale up predictive models, assuming the data is available to mine.They also developed a way to show clinicians what exact data its model “looked at” for each patient it predicted an outcome for.
The accuracy of the Google AI in predicting whether a patient will die within 24 hours after admission is as high as 95 percent, which is 10 percent more accurate than models that the healthcare industry is currently using.
The AI is able to analyze previously out of reach data, such as notes scribbled on charts or buried in PDFs. The neural network had no trouble digesting all the information to make the best predictions possible, which came faster and more accurate compared to traditional methods.
Business vs A.I. in Healthcare
Systems like these raise many ethical questions, though. Let’s move a bit forward into the future. Try to imagine a patient who is waiting for his cancer treatment in hospital. If his insurance company would have access to the predictive system and they would already know that he is going to die within a couple of months with 95% accuracy, do you really think they are going to cover the 40k USD treatment bill? Or will they rather prefer some cheaper treatment solution? I personally believe that governments all around the world in the future will have to address this issue in legislation.
Stigma still surrounds artificial intelligence, due to the technology’s portrayal in science fiction movies and recent developments such as the creation of the world’s first psychopath AI. However, artificial intelligence has a lot to offer to the healthcare industry, with the potential to save thousands of lives in the future.
What do you think about insurance companies having an access to predictive systems like these? Let us know in the comment.
Follow us here https://twitter.com/GREYHAWKAI1
Check out our BlockDelta profile for our full contact details.