> Background: The implementation and clinical impact of machine learning–based early warning systems for patient deterioration in hospitals have not been well described. We sought to describe the implementation and evaluation of a multifaceted, real-time, machine learning–based early warning system for patient deterioration used in the general internal medicine (GIM) unit of an academic medical centre.
> Methods: In this nonrandomized, controlled study, we evaluated the association between the implementation of a machine learning–based early warning system and clinical outcomes
> Results: The study included 13 649 patient admissions in GIM and 8470 patient admissions in subspecialty units. Non-palliative deaths were significantly lower in the intervention period than the pre-intervention period among patients in GIM.
NeurogenesisWizard on
Seems like an ok use of ai unless of course they are not going to deteroriate due to alternate unique metabolisms, then they get poisoned upon evaluation. Because ai has more statistics on the more common traits, people with alternate metabolisms would be more at risk. Albeit rare. But depends on the data used. And was said data obtained ethically? Well, some things aren’t worth agonizing over.
-SandorClegane- on
So…does it keep patients from dying or simply offer more realistic expectations of the deaths that already occur?
Academic_Signature_9 on
Small world.
I think I met one of the authors of this study at an AI / Big Data conference a while back. They were worrying about the ethical issues behind data collection and whether or not they’d be able to continue.
Looks like it worked out.
aedes on
Clinical studies that use a pre- and post-intervention design in the COVID era are usually uninterpretable, unfortunately.
> The GIM intervention cohort included all patients cared for by a GIM team on the GIM unit between Nov. 1, 2020, and June 1, 2022…
>The GIM pre-intervention cohort included all patients cared for by a GIM team on the GIM unit between Nov. 1, 2016, and June 1, 2020.
Essentially, their pre-intervention mortality rate will be artificially raised by COVID deaths in the first wave before immunizations and treatments were available.
5 Comments
> Background: The implementation and clinical impact of machine learning–based early warning systems for patient deterioration in hospitals have not been well described. We sought to describe the implementation and evaluation of a multifaceted, real-time, machine learning–based early warning system for patient deterioration used in the general internal medicine (GIM) unit of an academic medical centre.
> Methods: In this nonrandomized, controlled study, we evaluated the association between the implementation of a machine learning–based early warning system and clinical outcomes
> Results: The study included 13 649 patient admissions in GIM and 8470 patient admissions in subspecialty units. Non-palliative deaths were significantly lower in the intervention period than the pre-intervention period among patients in GIM.
Seems like an ok use of ai unless of course they are not going to deteroriate due to alternate unique metabolisms, then they get poisoned upon evaluation. Because ai has more statistics on the more common traits, people with alternate metabolisms would be more at risk. Albeit rare. But depends on the data used. And was said data obtained ethically? Well, some things aren’t worth agonizing over.
So…does it keep patients from dying or simply offer more realistic expectations of the deaths that already occur?
Small world.
I think I met one of the authors of this study at an AI / Big Data conference a while back. They were worrying about the ethical issues behind data collection and whether or not they’d be able to continue.
Looks like it worked out.
Clinical studies that use a pre- and post-intervention design in the COVID era are usually uninterpretable, unfortunately.
> The GIM intervention cohort included all patients cared for by a GIM team on the GIM unit between Nov. 1, 2020, and June 1, 2022…
>The GIM pre-intervention cohort included all patients cared for by a GIM team on the GIM unit between Nov. 1, 2016, and June 1, 2020.
Essentially, their pre-intervention mortality rate will be artificially raised by COVID deaths in the first wave before immunizations and treatments were available.