remember in the early noughties when everyone was obsessed with those online death calculators? that bit of mildly entertaining (and extremely morbid) tech was popular but it’s safe to say it was based on some random algorithm. but of course, stanford researches have added a dash of artificial intelligence (AI), and now there’s an ‘accurate’ system to predict the end of our lives…

 

to develop the technology, stanford researchers trained a deep neural network using 2 million hospital records. as a result of collecting all this data it claims to be able to predict when a patient will die with up to 90 percent accuracy. the team behind the work says it could vastly improve end-of-life care for patients and their families, by more accurately pinpointing when a terminal or seriously ill patient may pass.

 

in the new study published pre-print to arXiv, the stanford team explain that their AI algorithms rely upon deep learning, a machine learning technique that uses neural networks to filter and learn from huge amounts of data. this allows it to build what can be called an ‘all-cause mortality prediction model’, instead of being disease or demographic specific.

 

the criteria for deciding which patients benefit from palliative care can be hard to state explicitly,’ the authors explain in the paper. our approach uses deep learning to screen patients admitted to the hospital to identify those who are most likely to have palliative care needs. the algorithm addresses a proxy problem – to predict the mortality of a given patient within the next 12 months – and use that prediction for making recommendations for palliative care referral.’