Artificial Intelligence to improve medical image analysis

In the U.S. alone, between 6 and 7 million people break a bone every year. Even though this is a common injury, it isn’t a trivial injury. Even when the fracture heals smoothly and without any complications, the treatment still requires an investment in terms of money, time, and resources. Moreover, the best physician may misinterpret X-ray results. If we sum up all these factors, we clearly see how fractures are, in fact, a big deal.

What if we were to take a fresh look at how we diagnose bone fractures? What if we could use an assistant and have a second opinion in a matter of seconds? And believe me, the assistant I’m suggesting is a meticulous one: it’ll never overlook anything. And what if this assistant could speed up the diagnostic process and reduce healthcare costs as a result?

I bet you would be interested in having a look at its resume. It doesn’t have one, though, as it is a software. But beware, I’m not talking about traditional Computed Assisted Diagnostic (CADx). What I’m suggesting goes beyond that. This machine has a brain and has received training on medical image interpretation. Welcome to the era of cognitive computing!

Are you wondering how a machine can learn to read an MRI or a radiography? In the same way radiologists do: building a repertoire of images together with their diagnoses. In this way, real and virtual MDs establish the features of a healthy bone or a damaged one. Stating that the cognitive process is the same, doesn’t mean that a software will replace healthcare practitioners. Health professionals are the people with the ultimate responsibility to come up with a diagnosis. Nevertheless, a second pair of eyes that can make your job easier, spot even less noticeable abnormalities, and reduce hospital discharge times. A good investment for patients, hospital staff, and health insurance companies, wouldn’t you agree?

About Author

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>