“Hopefully they’ll catch it, but what if they miss?” said Dr. Harpal Sandhu, an assistant professor of ophthalmology at the University of Louisville.

That was the inspiration behind new research done by Sandhu and colleagues at UofL, which aims to make the process of identifying diabetic retinopathy more objective by teaching software to diagnose. The results were published in the British Journal of Ophthalmology and the .

The team used a technique called machine learning to teach the software how to identify the condition, which occurs when high blood sugar levels cause damage to the blood vessels in the eye.

They showed the software pictures of eyes — tons of them, both healthy and unhealthy — until it “learned” what to look for when making a diagnosis. It learned to look for swollen vessels or leaking of blood into the eye, which could cause impaired or even lost vision.

When tested on more than 100 cases, the software got 94.3 percent of the diagnoses correct.

Sandhu said diagnosis of diabetic retinopathy is especially important, given the “epidemic proportions” of diabetes. According to the , the number of people with diabetes rose from 108 million in 1980, to 422 million in 2014.

The software combines two systems developed by the UofL team for optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA). Together, they allow the machine to, non-invasively, get a clear view of the blood vessels in the eye and make a diagnosis.

The software works with a machine used by many optometrists and ophthalmologists to examine patients’ eyes. The automated technology could supplement care in areas where medical staffing is low and optical specialists are sometimes unavailable.

“We think this machine could be set up in the family doctor’s clinic,” said Dr. Ayman El-Baz, professor and chair of bioengineering, who was part of the development team.

He said the technology is part of a “new trend for using artificial intelligence in medicine.” Similar research is being done at UofL for diagnosis of other medical conditions, such as lung cancer and organ rejection.

The development was backed by from the Wallace H. Coulter Translational Partnership, which promotes collaborative translational research, and is currently available for through the UofL Office of Technology Transfer.

This project is “a great example of collaboration between engineers and clinicians to address unmet clinical needs,” said Jessica Sharon, director of the Coulter Translational Partnership at UofL, adding the resulting technology “has the potential to make a significant impact through commercialization to diagnose disease and improve patient outcomes.”

Typically, adult neurons cannot make new synaptic connections as easily as developing neurons. That limits the potential for recovery from injury to the brain and spinal cord. One type of neurons in the retinas of mice, OFF-type retinal bipolar cells, has the unusual ability to make new connections into adulthood. Under normal conditions however, these cells only develop new connections with a few cells and within a limited area known as a tile. The function of these cells is to receive information from photoreceptor cells and send it along the optic nerve to the brain.

“These neurons continue to develop and elaborate their connections within their established group of cone cells in the retina,” Borghuis said. “This suggests synaptic plasticity, or the ability for the neurons to create new connections with other neurons. This is significant because in brain disease, you want to transplant and regenerate neurons and integrate them through the formation of new synapses with other neurons.”

In the first stage of the work, a team of researchers at the University of Idaho led by Peter Fuerst, PhD, determined that removing the gene encoding a protein known as Down syndrome cell-adhesion molecule (DSCAM) allows these cells to extend neuronal connections beyond their normal tile barriers. They genetically modified the mice to omit DSCAM from those cells, after which the cells were seen to form apparent contacts with neurons outside their tiles.

However, the researchers were unable to determine whether those apparent neural connections were, in fact, functional and capable of transmitting visual information.

That’s where Borghuis comes in. Using a unique imaging and recording technique pioneered in his laboratory at UofL, two-photon fluorescence-guided electrophysiology in deep layers of the neural retina, Borghuis recorded the bipolar cells’ responses to visual stimulation. His measurements showed enlarged visual receptive fields in the genetically manipulated retinal neurons, demonstrating that the extended cells made new, functional synapses onto cones.

“Right off the bat we could see that the receptive fields were larger, so we could tell that their visual responses were consistent with neural outgrowth and new synapse formation,” Borghuis said.

These tests proved the neural outgrowth seen by the Idaho team led to stable, functional connections with new cells.

A new joint research grant from the National Institutes of Health, awarded equally to Borghuis and Fuerst, will fund the collaborative research for another two years. During that time, they will induce the DSCAM knockout later in the lifespan to determine the identity and strength of the new synapses. In addition, they also will perform studies of neurons at later synaptic stages within the retina to determine other potential consequences of increased neuron growth at the level of the visual input.

“We have known about the tiling or mosaic structure of these cells for decades, and there are models and ideas for why neurons should tile. Now that we have a genetic tool that allows us to disrupt tiling within a neuron population experimentally, we can finally test these models,” Borghuis said.

The ability to stimulate neural outgrowth with new synaptic connections may ultimately improve humans’ ability to recover from brain and spinal cord injury or disease by supplying new neural connections. Even more promising, it could lead to neural regeneration and transplantation-based therapies for restoring visual function in retinal diseases such as diabetic retinopathy and macular degeneration.