AR AI model can objectively identify, locate pain in real time

Register now

Machine learning, combined with neuroimaging data, has the potential to objectively determine whether a patient is suffering pain and where it’s located.

Accurate pain assessment is critical to provide proper diagnosis and treatment, medical experts agree. However, it’s difficult to quantify pain, and most assessments are subjective. Subjective assessments are inconsistent and can’t be used when a patient can’t communicate, such as during surgery. They’re also of limited value in understanding the neurophysiological processes underlying different types of pain.

The study authors, from the University of Michigan and elsewhere, tested the feasibility of a prototype of a mobile neuroimaging-based clinical artificial reality and artificial intelligence (CLARAi) framework for objective pain detection and localization in the clinical environment.

The portable CLARAi platform combines visualization with brain data using neuroimaging to navigate through a patient’s brain while the patient is sitting in the facility.

The researchers triggered dental pain by administering cold to the teeth of 21 volunteer participants with hypersensitive teeth. They used a portable optical neuroimaging technology, functional near-infrared spectroscopy, to detect and collect the participants’ cortical blood flow and oxygenation activity during the evoked clinical pain, thus measuring brain activity and responses to the pain.

They used the brain pain data to develop algorithms to predict pain or its absence.

Wearing Microsoft’s special augmented reality glasses, called HoloLens, the researchers viewed the participant’s brain activity in real time on a reconstructed brain template, while the participant sat in the clinical chair. The red and blue dots on the image showed the location and level of brain activity, and this “pain signature” was mirror-displayed on the augmented reality screen.

The algorithms, used with internally developed software and the neuroimaging data, predicted pain or the absence of pain about 70 percent of the time.

“In a true clinical environment, with such framework, clinicians can better understand in an objective way to determine when/where the patients are suffering from pain, especially when they cannot express themselves,” the researchers stated.

The study was published in the Journal of Medical Internet Research.

“Although extensive validation work still needs to be done, the CLARAi framework might turn into reality the goal of precisely “seeing and believing” the biologic pain suffering of our patients in the doctor’s office,” the authors concluded.

For reprint and licensing requests for this article, click here.