Dec 15, 2017
Merriam Webster defines AI as, “1 : a branch of computer science dealing with the simulation of intelligent behavior in computers. 2 : the capability of a machine to imitate intelligent human behavior”. So are we there yet in radiology where AI is capable of truly intelligent medical image analysis or is it going to steal human jobs?
If you think artificial intelligence is already available in Healthcare or way out there in the distant future, consider the headline from earlier this year, First FDA Approval For Clinical Cloud-Based Deep Learning In Healthcare. This application available in 2017 is narrowly focused on cardiac analysis where it claims manifold improvement with results in 15 seconds compared to a professional human analyst taking 30 minutes! Such speed leads to consistent efficiency towards a faster diagnosis.
A study identified that machine learning using a Breast AI algorithm could reduce unnecessary surgery by 30%. This presents a more balanced approach to guide radiologists to avoid recommending unnecessary interventions where recently a Vast Study Casts Doubts on Value of Mammograms. So now we have accuracy with further reduction of costs by altogether avoiding interventions with better patient experience. All this is possible while assisting radiologists who can now focus on the larger picture beyond the dedicated screen time in dark rooms.
There are many things to focus on, starting from simply ensuring that the radiologists diagnosis is actually read and discussed with the referring physician. Next could be the radiologist stepping out in the daylight and getting some real face time with patients to better inform them of the diagnosis. Then there is the whole field of interventional radiology in the operating room to guide critical surgeries in real time to ensure the best results.
I believe that AI offers a path to drive meaningful value in radiology instead of being viewed as a cost center. I think it would be interesting to investigate and compile what AI based products exist in radiology today and where could this arc lead us.