A backlog of work has become the norm in many NHS trusts around the country, and many patients are left for weeks before they can get their results from a radiological study. Recently there has been a buzz around the use of Artificial Intelligence in Radiology, and how it could potentially be used to shorten waiting times, but just how viable is this?
The good news is there has already been significant progress.
Machine learning has advanced hugely over the last decade, and this has opened up a number of new avenues for companies to explore. One exciting company based in the United States called Enlitic is in the process of bringing a new product to market that they claim reduces the error rates of Radiologists, as well as speeding up their work. Entilics software is simple, and easy to use, firstly an image is loaded on to the platform and assessed for abnormalities. If anything is found it is highlighted, and the images are given a high priority status and brought to the attention of an appropriate Radiologist. The software will even auto generate most of the report for the Radiologist to get rid of one of the most tedious parts of the job. The software’s aim isn’t to get rid of Radiologists, but to improve and speed up their work.
Another company based in San Diego, USA called Cure Matrix has used its software to evaluate more than 500,000 mammograms. By doing this they claim their software can identify potential false negatives, which are undiagnosed cancers, and false positives, which are unnecessary recalls or patients that turn out to be normal. Again, this company claims it can be used in conjunction with Radiologists to improve patient outcome and the patient’s experience.
The next company is called Arterys, also from the USA. They’ve produced software, which can be used to assess images of the Lungs, Heart and Liver. But what AI commentators are most excited about is their 4D Flow technology, this programme is used in partnership with an MRI scan. MRI images of the heart are captured as usual and then the Artery software will generate precise measurements of blood flow, check for any abnormalities or disruption. Using a number of different tools it allows physicians the ability to evaluate valvular, and congenital heart disease. It also easily visualises and quantifies any shunts.
There are countless other AI companies that are hoping to break in to the huge medical sector all across the world, but I don’t think the end goal is to completely get rid of Radiologists. These programmes are designed to be used alongside Doctors to improve and speed up their work, and from the clinical trials it certainly seems to be having the desired effect. These AI programmes should be embraced, and put in place quickly, especially when we consider the shortage of Radiologists we have in the UK. If the demand for imaging keeps on going up we need to find new and innovative ways to keep up with the demand.
What are your thoughts about the use of Artificial Intelligence for healthcare?