The Guardian
Email YouTube Facebook Instagram Twitter
Everything you need to live well

Scientists Develop Al That Can Predict When You Will Die

By Njideka Agbo 17 February 2018   |   1:00 pm

Researchers at Stanford University have developed an Artificial Intelligence system that can predict the death of hospital patients. Their findings were published in a paper titled, “Improving Palliative Care with Deep Learning”.

The computer system Al predicted the death of over 40,000 hospital patients between three to twelve months with 90% accuracy.

According to IEEE Spectrum, they used the records of about two million patients (adult and child) from Electronic Health Record data from the Stanford Hospital and Lucile Packard Children’s hospital.

As Anand Avati, a member of the AI Lab at Stanford University, explained, “The scale of data available allowed us to build an all-cause mortality prediction model, instead of being disease or demographic specific.”

Although it is a pilot test, the scientists hope that it will aid in understanding and improve healthcare especially when it concerns people terminal diseases after more data is analyzed.

Kenneth Jung, a research scientist at Stanford University, believes in the future of the AL in medical practice:

“We think that keeping a doctor in the loop and thinking of this as ‘machine learning plus the doctor’ is the way to go as opposed to blindly doing medical interventions based on algorithms… that puts us on firmer ground both ethically and safety-wise”.