Artificial Intelligence

Imagine this …. you wake up as usual, start the coffee, give your beloved Fido a treat and hop in the shower. You dress and head to the office. Upon arrival, you chat briefly with colleagues, turn on the computer, make a second cup of coffee, and review your schedule and projects. It’s a normal day and you’re already thinking about the weekend plans you have. Suddenly, you feel a pressure sensation in your chest that must be that annoying indigestion. You chew a few Tums, drink some water, and try to focus on work. After all, it’s just indigestion. You start to feel a little short of breath and a sense that something just isn’t right. A colleague stops by and says you don’t look well. You nod and say it's indigestion, but you can’t shake the sense of increased tightness in your chest.   Why aren’t the Tums working?

Another co-worker stops by and mentions that her father had a heart attack at age 47 and says she thinks you need to go to the hospital. You protest, but she has now informed your manager who insists on calling 911.  You arrive to the ED and after radiographic studies, laboratory testing, and an EKG, you are admitted overnight for observation. Fast forward to the morning, and per the doctor, your “heart tests look just fine”.  No heart attack the doctor says, likely reflux and prescribes some medication with recommendations to have a stress test in the outpatient setting. Time to go home.

Two weeks later you’re at work and suddenly feel lightheaded, have noted extreme thirst all day despite drinking copious amounts of water and you unexpectedly pass out. Back to the ED you go. Upon arrival, your glucose (sugar) level is > 500. Much to your shock you are diagnosed with Diabetes Mellitus Type II.

A review of labs obtained during your last admission by the current doctor reveals your glucose (blood sugar) levels were in the 160’s which is well above the normal level.  Had it been noticed before your discharge you could have been treated then and avoided the new admission.  Why didn’t the doctor notice the blood sugar levels before?

This is where AI may have been beneficial? Medical professionals may be well educated, but they are still human. With the ever-increasing demands in healthcare settings, patients are lucky if the clinician can spend 10 minutes for an examination and review of history, risk factors, and what can seem like an infinite list of lab results. In these times of increased health care costs, the move towards “value-based” medicine, cost restraints, and insurance restrictions how can one clinician be tasked with attending to ever-increasing patient loads, scan all your tests before you go home to make sure nothing is missed? Would a way to alert your healthcare provider, not just the red highlighted numbers that are eye exhausting, be of benefit?

According to SAS.com, “The term artificial intelligence was coined in 1956, but AI has become more popular today thanks to increased data volumes, advanced algorithms, and improvements in computing power and storage." Are you involved with AI? Do you have technology or know of technology that could have helped this patient? Would love to hear your thoughts!