Dive Brief:

• Optina Diagnostics said it has received FDA breakthrough device designation for a retinal imaging platform that uses artificial intelligence to detect amyloid plaques in the brain, a hallmark of Alzheimer's disease.
• The eye-imaging test is intended as a non-invasive, low cost tool to improve diagnostic accuracy for patients who are being evaluated for Alzheimer's and other forms of cognitive decline.
• The platform is being developed as an adjunct to other diagnostic assessments, the Montreal-based company said.

Dive Insight:

Optina Diagnostics said its platform uses AI to analyze hyperspectral retinal images captured in a simple eye scan. The company said its noninvasive test is intended as an aid in the detection of likely positron emission tomography (PET) amyloid status in patients who are undergoing evaluation for cognitive impairment.

FDA's breakthrough device program is intended to speed development and review of medical devices that offer a more effective treatment for life-threatening or irreversible debilitating diseases.

Optina said that by 2050 as many as 14 million people in the United States may be affected by Alzheimer's, a fatal disease with no current treatment to delay its progression. It is the sixth leading cause of death in the United States, according to the National Institute on Aging.

Symptoms such as problems with memory and thinking skills typically first appear in people in their mid-60s. The disease is characterized by abnormal deposits of proteins from amyloid plaques and tau tangles in the brain that kill neurons or stop them from functioning.

PET scans use an injection of a radioactive substance called a tracer to measure brain activity such as glucose use. Because amyloid PET is an expensive technique of limited accessibility, the diagnosis of Alzheimer's disease in clinical practice is typically based on clinical symptoms, Optina said.

The company said its eye imaging test can improve diagnosis accuracy and management of patients for Alzheimer's and other forms of cognitive decline.

The technology also can be applied to accelerate and reduce the cost of enrollment of patients in clinical trials for Alzheimer's disease treatments, the company said.

Diabetic eye diseases have been an early focus in the development of AI-based diagnostic tests involving retinal imaging. FDA last year approved the first autonomous AI diagnostic system cleared for sales in the United States, IDx-DR's software. That software detects diabetic retinopathy in images taken by retinal cameras. Alphabet's Verily unit is also working in this area, with a program underway in India.

In April, FDA released a proposed framework for determining what types of AI-based software as a medical device (SaMD) could be exempted from premarket submission requirements. The discussion paper, which suggests a total product lifecycle approach that includes performance monitoring, is part of the agency's effort to adapt to emerging technologies.