A lot of AI in healthcare use cases require lengthy and time-intensive integration, a complete overhaul of workflows, and retraining of employees. This is not the case with Augmented Reality, however, and when you pair the technology with AI, the implications for the industry, while still uncommon are quite significant.
AR is much more like traditional software solutions, in that it can be implemented into workflows without much disturbance. Google Glass, for example, may have struck out with consumers but the tech has for sure piqued the interest of physicians. Google Glass allows doctors to view a patient's notes without taking their eyes off the patient, thus allowing physicians to be more present. While this is still some way away from being rolled out on any sort of scale, it does offer a glimpse into the future of AI-enhanced AR in Healthcare.
Two solutions, in particular, have already made significant strides in the healthcare industry.
The first, purports its software can reduce the number of times a nurse or physician sticks a patient with a needle to find a vein.
While the second, claims its software can visualize 3D images of a patient’s organs and allow doctors to interact with those images.
Let's dive deeper into these technologies...
Taking the guesswork out of finding a vein...
AccuVein helps healthcare professionals to find the best vein possible for IV and blood draws. Vein visualisation uses Near-infrared (NIR) imaging for detecting veins, increasing the likelihood of a successful first stick by 3.5 times in some cases.
How does it work?
The two lasers work in tandem to provide a real-time image of the subcutaneous vasculature up to 10mm deep. First, the haemoglobin in the blood absorbs the infrared light, so there is a reduced amount of light reflection from the veins. Then, a custom detection system uses this change in reflection to determine vein location and pattern, which it digitally projects on the surface of the skin. Lastly, the red laser makes the vein mapping visible to the clinician.
The result is a visual projection that has centre line accuracy of less than the width of a human hair.
Does it work?
AccuVein claims to have helped four haemophilia treatment centres in France reduce the cases of difficult venous access (DVA) among patients. The study involved 450 participants. Among the total participants, access to the veins was reported to be difficult in 165 because they had poor vein condition, were too young, or were overweight. Of this number, actual difficulty in locating veins was encountered in 82.4%, and a fourth of these patients needed more than one puncture attempt.
So while this might seem insignificant to some, it may not be for someone with a chronic illness or a sick child.
A 3D medical visualisation software that allows healthcare professionals to visualise and interact with 3D images that depict human tissue and organs.
The main application of this technology is as a pre-surgery software to simulate and evaluate surgical treatment options.
Another application is in pediatric cardiology, 3-D imaging allows surgeons to do a run-through on a life-size patient replica, reducing the guesswork in interpreting images. That, in turn, has reduced the surgical time from about four hours to as little as 90 minutes
How does it work?
While MRI and CT scans create flat two-dimensional images, The True 3D Viewer provides high definition visual and interaction Computed Sensory Cues to depict surgical views in an open 3D space
Through machine learning, Echo Pixel changes the way doctors use an image and more importantly records the way other doctors have interacted with it previously. In this way, other doctors can follow the same steps and methodology.
“It really helps doctors to not think about the data. They can focus on the clinical issues.”- said Sergio Aguirre, EchoPixel's founder and chief technology officer.
Does it work?
Researchers found that, compared with traditional 2D tomography techniques, the images captured by the True 3D Viewer improved the surgeon’s ability to see and interpret anatomical details to 90% from 81% previously.
In addition, the time for a physician to interpret a tomographic readout declined significantly when using information from True 3D
Since the crazed trend of Pokemon Go ended, augmented reality has been quietly progressing and being put to more practical use. Together with AI innovation, AR is set to partner Machine Learning and Deep Learning (albeit at a lesser scale) in leading the healthcare revolution.