Does software have a purpose in clinical diagnostics?

I recently read an article in the New York Times about the use of software to develop differential diagnoses. It talks about Dr. Gurpreet Dhaliwal, a professor of clinical medicine at the University of California, who is considered to be one of the great diagnosticians of our time. He is able to come to a correct diagnosis through intricate thinking processes and experience, but he also gives credit to a diagnostic program that he uses as a “second check” called Isabel. Isabel is certainly not the only diagnostic software program on the market. In fact, the history of computer assisted diagnostics dates back to the 1970s, with multiple programs currently on the market.

And it’s not just internists who can use this type of software. Radiologists have also been developing computerized schemes that can automatically detect and quantitate abnormalities in radiological imaging. The purpose of it is to improve diagnostic accuracy and the consistency of a radiologist’s interpretation by using what the computer outputs as a guide or a second opinion.

My first instinct was to think about how useful this type of adjunct could be to our regular diagnostic processes. If we could successfully merge the consistency and preciseness of computer software into our own routine thinking processes, I would think that we might be less likely to miss diagnoses and more likely to consider diagnoses we might have otherwise disregarded.

On the other hand, computer assistance is unlikely to be adopted by all medical practitioners. Many Physicians use experience and pattern recognition to determine a diagnosis, and often do not need the aid of a computer to output a list of differential diagnoses. Furthermore, isn’t medicine more than just sifting through a list of possible illnesses? I think that Medicine is much more of an art than it is a science. It’s about understanding patients’ motivations, showing empathy and effectively communicating and collaborating.

I think at this point in time the use of computer assists is perhaps a little too simplistic. If Medicine could be boiled down to simply plugging in patient factors and outputting the likeliest diagnosis, we would all have been replaced by computers by now. Nonetheless, how and where these types of tools are incorporated into practice in the future remains to be seen.

Original article:

http://www.nytimes.com/2012/12/04/health/quest-to-eliminate-diagnostic-lapses.html?pagewanted=all&_r=1&

-Carolyn