Every day, clinicians around the world look at hundreds of thousands of medical test results.
Things like x-rays, blood test results, MRI scans and the like, all must be studied and interpreted by experts.

These experts have many years of experience in their fields and have become very good at interpreting test results; identifying the patients that need treatment, and how urgently.

Radiologists, the medical specialists that use medical imaging to diagnose and treat diseases can view thousands of x-rays a day. With their many years of experience in this field, they’ve become adept at identifying which images require immediate follow-up or intervention. Radiologists can determine, for example, whether the lungs of a trauma patient are collapsed or if there is internal bleeding that requires immediate intervention. They can identify if a patient has a bone fracture, and whether they need a cast or not.

What they can’t do is speed up the process or make it more efficient. The ability of radiologists to read the images and know what they’re seeing isn’t the main barrier to progress. The real challenge is that radiologists only have so many hours in a day. The speed at which they can accurately read and evaluate images can only be so fast. While brilliant, they’re not as efficient as a computer; they are only human after all. 

Many of the x-rays and test results analysed by doctors turn out to be normal. This means that patients who need the most urgent attention can have their results hidden amongst a stack of less than urgent results. Doctors have to spend their valuable time going through all the results to make sure each patient is treated fairly, meantime those with serious conditions can experience delays in receiving treatment. 

This problem exists throughout healthcare, from doctor’s offices to emergency rooms. The ability to have a machine read and report on test results would be of huge importance to the healthcare industry. Clinical decision support (CDS) is a process for enhancing health-related decisions, notifications and actions with pertinent, organised clinical knowledge and patient information to improve healthcare delivery. A machine learning platform capable of analysing test results and advising experts on the best course of action would be an example of a CDS system.

These machine learning programs can be trained with existing test results and programmed to identify those that are irregular. One way this is accomplished is by giving the program a database of 100 x-rays of normal healthy arm bones, and then show it various x-rays of fractured arms bones. The computer can then be trained to flag all future x-rays of arm bone fractures. The same method could be applied to any other test results such as MRIs or blood tests. This would not only save time for the human experts – as the normal results could be identified earlier – but it would be much faster, meaning at-risk patients would receive treatment as soon as possible. Then clinicians could spend more of their time with the at-risk patients, who would benefit from this extra intensive clinician time. 

CDS will have enormous benefit for the improvement of healthcare for individual patients, hospitals, and the health of the national population.


To learn more about Clinical Decision Support systems, the drivers and challenges involved and the basic CDS functionality required to start making positive change in the healthcare ecosystem, download the white paper now!