Duke University researchers have used AI to boost the resolution of optical coherence tomography (OCT) to improve medical images across fields, from cardiology to oncology. Their findings were recently published in nature photonics.
The OCT imaging technology is likened to an ultrasound that uses light instead of soundwaves, by shooting a probe of light into a tissue and basing the boundaries of the features within by the delays of the light waves as they bounce back.
The new technique, dubbed optical coherence refraction tomography (OCRT) improves upon the old methods of creating OCT images with high lateral resolution, which previously relied on holography and measures the complex electromagnetic field reflected back from the object.
“An historic issue with OCT is that the depth resolution is typically several times better than the lateral resolution,” lead author Joseph Izatt, of the Michael J. Fitzpatrick Professor of Engineering at Duke, said in a statement. “If the layers of imaged tissues happen to be horizontal, then they’re well defined in the scan. But to extend the full power of OCT for live imaging of tissues throughout the body, a method for overcoming the tradeoff between lateral resolution and depth of imaging was needed.”
The old OCT method requires the sample and imaging apparatus to be perfectly still.
“This has been achieved in a laboratory setting,” said Izatt, who also holds an appointment in ophthalmology at the Duke University School of Medicine. “But it is very difficult to achieve in living tissues because they live, breathe, flow and change.”
The new technique takes OCT images from multiple angles to extend the depth resolution to the lateral dimension. And Izatt and his team has to accurately model how the light was bent as it passed through the sample and compensate for altered paths of light refraction. They leveraged machine learning tools to develop a method to determine light directions and create a better image.
In a proof-of-concept experiment, researchers used bladder and trachea tissue samples from a mouse in a tube that was rotated beneath an OCT scanner. The algorithm boosted the lateral resolution of the images more than 300% and reduced the background noise from the final image.
“Capturing high-resolution images of the conventional outflow tissues in the eye is a long-sought-after goal in ophthalmology,” Sina Farsiu, of the Paul Ruffin Scarborough Associate Professor of Engineering at Duke, said when referring to the eye’s aqueous humor drainage system. “Having an OCT scanner with this type of lateral resolution would be very important for early diagnosis and finding new therapeutic targets for glaucoma.”