Engineers from Duke University have harnessed the power of machine learning to increase the resolution of optical coherence tomography (OCT) imaging, according to an Aug. 19 study published in Nature Photonics.
The new method, dubbed optical coherence refraction tomography (OCRT), improved resolution down to a single micrometer and could ultimately enhance medical images used in fields from cardiology to oncology, corresponding author Michael J. Fitzpatrick, a professor of engineering at Duke, said in a news release.
"An historic issue with OCT is that the depth resolution is typically several times better than the lateral resolution," Fitzpatrick added. "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."
For their new technique, the researchers utilized OCT images gathered from multiple angles to extend the depth resolution to the lateral dimension, but each image became blurry as light refracted through irregularities in cell and tissue samples. They turned to a machine learning expert who developed an algorithm of the tissue’s refractive index that took the light distortion into account.
Fitzpatrick and colleagues tested their algorithm in a proof-of-concept experiment. They placed bladder or trachea tissue samples taken from a mouse into a tube and rotated them beneath an OCT scanner. The algorithm created a map of each sample’s refractive index, increasing the lateral resolution of the image more than 300% while limiting background noise.
“OCT has already revolutionized ophthalmic diagnostics by advancing noninvasive microscopic imaging of the living human retina," said co-author Joseph A. Izatt, a Duke University engineer. "We believe that with further advances such as OCRT, the high impact of this technology may be extended not only to additional ophthalmic diagnostics, but to imaging of pathologies in tissues accessible by endoscopes, catheters and bronchoscopes throughout the body."