Imagine the future when you go to your doctor to be treated for some sort of ailment. After collecting and entering key data into an app, she’s presented with a clinical diagnosis and optimal course of treatment, including the correct medications. While she’s there to talk “human-to-human” to the patient, it’s a predictive model and algorithm that’s done all the heavy lifting. That day is rapidly approaching, if not already here. Are we ready for it?
Everywhere we look these days, someone’s talking about Artificial Intelligence in Healthcare (AI). It started with ads from large computer manufacturers and technology firms. The list of AI advancements and new capabilities is almost too long to list. AI knows no boundaries, and it has recently been introduced to our National Pastime. You may have read that the Atlantic League, an eight-team independent minor league baseball association is experimenting with software to call balls and strikes. The umpire (human) is there for quality control purposes only, and to verbally say ball or strike. A ball that hits the ground in front of home plate and bounces through the strike zone is not a strike!
The healthcare Industry is now fully engaged in utilizing AI and machine learning to assist with the treatment of patients. Earlier this year, CMS announced the Artificial Intelligence Health Outcomes Challenge. The primary focus is on demonstrating how AI tools can be used to predict unplanned hospital and skilled nursing facility admissions and adverse events. Proposals were due in mid-June, with an anticipated announcement of Stage 1 participants by July 19. It’s fairly certain that no one could have predicted the level of interest in this challenge. As it turns out, CMS received over 300 applications, and they are still in the process of determining the Stage 1 participants.
The impact of AI in healthcare can be seen everywhere. Currently in the testing phase, HealthEC expects to unveil a solution soon that will identify key factors that influence hospital readmission for diabetes and predict the likelihood of diabetic patient readmission. With preliminary results showing the solution being more than 90 percent accurate, we are very enthusiastic about its future success. AI is also being used to determine the right drug(s) to be given to a patient through pharmacogenomics, where a patient’s genomic data can provide an accurate prediction of the most effective medication for a patient’s condition.
Assuming that AI tools become readily available, the real question is whether physicians will view this advancement as something helpful, or something that poses a threat to their livelihood and can’t be trusted to replace the number of years of education and training required to become a licensed physician. The answer to this question will determine whether AI will have a meaningful future in the doctor’s office. What do you think?