Donate to Science & Enterprise

S&E on Mastodon

S&E on LinkedIn

S&E on Flipboard

Please share Science & Enterprise

Augmented Reality, A.I. Harnessed to Visualize Pain

Dental work


28 June 2019. Researchers at University of Michigan created a system with real-time neurological imaging, algorithms, and augmented reality to detect and visualize a patient’s pain. A team from the university’s dentistry school in Ann Arbor describe the workings and feasibility tests of the system in today’s issue of the Journal of Medical Internet Research.

The paper’s senior author, dental materials science professor Alexandre DaSilva and co-author Eric Maslowski are founders of the spin-off company MoxyTech Inc. in Ann Arbor to commercialize the technology. Maslowski is MoxyTech’s president and DaSilva is the company’s chief scientist.

DaSilva and colleagues are seeking more objective methods for patients to describe the pain they’re experiencing to clinicians. At present, patients can express the extent of their pain on subjective rating scales, which many physicians and dentists find insufficient. In addition, people under anesthetics or with speech difficulties may not be able to report their pain experience.

“It’s very hard for us to measure and express our pain, including its expectation and associated anxiety,” says DaSilva in a university statement. “Right now, we have a one to 10 rating system, but that’s far from a reliable and objective pain measurement.”

The team from Michigan’s Headache and Orofacial Pain Effort, or HOPE, lab, with associates from MoxyTech and consumer products company Colgate Palmolive that funded the research, developed the system. The HOPE lab studies pain occurring in the face and head from chronic disorders of the trigeminal nerve, a large cranial nerve sending sensory and motor information to the face, nose, sinuses, and jaw.

The system the researchers call clinical augmented reality and artificial intelligence, or CLARAi, captures data on a patient’s brain activity indicating pain experiences. The patient wears a head-covering cap with sensors that read blood flow and oxygenation, indicators of signals received in areas of the brain associated with pain. The sensors use near infrared spectroscopy to interpret the signals and transmit the data to a computer for analysis.

The data are assembled into images indicating the number and extent of brain signals in real time, which are then interpreted with a convolutional neural network that combines features of image analysis and machine learning. In a convolutional neural network, an algorithm dissects an image by layers to understand the features in the image. Different aspects of each layer discovered and analyzed by the algorithm are translated into data that the algorithm then uses to train its understanding of the problem being solved, with that understanding enhanced and refined as more images and data are encountered.

In addition, the analyzed data are sent and imaged in real time on Microsoft HoloLens augmented reality headsets. The headset displays a 3-D template of the patient’s brain, with a pain signature calculation also shown in the field of view.

The researchers tested CLARAi with 21 volunteers, who submitted to pain induced from 20 shots of cold air blown on their teeth. The results show the system accurately identifies pain or no-pain conditions about 80 percent of the time, with a three-layer convolutional neural network. Using a more sophisticated six-layer network, the team was able to accurately discriminate about 74 percent of the time between pain occurring on the left or right side of the head, as well as from a state of no pain.

While the tests prove the concept of CLARAi, the researchers plan to extend the technology to visualize pain in other parts of the body and develop a mobile version. University of Michigan owns the intellectual property rights for CLARAi, to which MoxyTech has a licensing option. The company now offers a smartphone app called GeoPain for patients to highlight specific areas of the body where they’re experiencing pain.

More from Science & Enterprise:

*     *     *

1 comment to Augmented Reality, A.I. Harnessed to Visualize Pain