Skip to navigationSkip to content
Close
AI Is Good (Perhaps Too Good) at Predicting Who Will Die Prematurely

AI Is Good (Perhaps Too Good) at Predicting Who Will Die Prematurely

Read more on Live Science

Contributions

  • This is the same work that population scientists have been doing with demographers and actuaries for decades. Not to belittle any of the work that has been done by researchers in training these data sets, the primary breakthrough that is yielding the results is not as much the heuristic algorithms, which

    This is the same work that population scientists have been doing with demographers and actuaries for decades. Not to belittle any of the work that has been done by researchers in training these data sets, the primary breakthrough that is yielding the results is not as much the heuristic algorithms, which have been actually used by statisticians in the past, but actually the advances in cloud computing allowing for massive amounts of processing power on tap. The real challenge for the industry has been having the processing power to be able to do the research. Cloud computing makes this a possibility.

  • Being able to more accurately predict surges in health degradation across population groups means we can counteract poor lifestyle habits more effectively and proactively plan and avoid bottlenecks in our medical supply chains and hospital capacity crunches that stifle the system right now as well as

    Being able to more accurately predict surges in health degradation across population groups means we can counteract poor lifestyle habits more effectively and proactively plan and avoid bottlenecks in our medical supply chains and hospital capacity crunches that stifle the system right now as well as using col encouraging . Encouraging!

  • "All three models determined that factors such as age, gender, smoking history and a prior cancer diagnosis were top variables for assessing the likelihood of a person's early death.

    The deep-learning algorithm delivered the most accurate predictions, correctly identifying 76% of subjects who died

    "All three models determined that factors such as age, gender, smoking history and a prior cancer diagnosis were top variables for assessing the likelihood of a person's early death.

    The deep-learning algorithm delivered the most accurate predictions, correctly identifying 76% of subjects who died"

    This could help insurance companies provide more customized (hence cheaper) life insurance products, rewarding healthier lifestyles, similar to what Progressive offers today but for car insurance with its Snapshot product, which personalizes your rate based on your actual driving: "The safer you drive, the more you save"

    #healthsciences #AI #emergingtech