Communicating the value of artificial intelligence (AI) in medicine is all about stats and figures. Statistics are a great way to explain the potential for AI in the healthcare system in terms that your busy CIO will understand.
Find below some of the standout stats and figures which were shared at AIMed 2017 in the workshop sessions and in the keynote talks at conference. For further reference, you can watch full videos of 2017 here.
Studies show that physicians underestimate the probability of negative outcomes for patients. A problem with prescriptive analytics for physicians is that if the tool points towards a positive outcome they provide confirmation bias for the physician, who is more likely to overestimate the probability of a positive outcome.
The average US patient can expect to be harmed by a diagnostic error at some point in their lifetime.
Although clinicians are working extremely hard around the clock, a lack of decision support for intelligence based decisions leads to a staggering variability in quality of care.
Physicians still rely on the ‘art’ of medicine, but AI will empower them to make decisions which are supported by a wealth of information about their patients and their diseases.
Hospital error is the third leading reason for patient death.
The healthcare industry needs to harness AI in medicine to predict these errors and protect patients.
Only 20% of medical decisions are made on the basis of results from randomised control trials (RCT), which are expensive to run.
Stat five and stat six
400,000 Americans are harmed by preventable errors.
By 2030 a single cognitive system will be able to generate more new biomedical knowledge in one hour than in the entire history of human scientific endeavour.
For more stats and also feature length deep dive articles, make sure to download your free digital copy of AIMed Magazine issue 01. It’s available here.