Researchers from University of California – San Diego have demonstrated how applying artificial intelligence together with machine learning techniques can screen patients with blinding retinal diseases. Through the use of a new computational tool diagnoses and treatment and be undertaken faster, to the benefit of the patient.
The machine learning was undertaken using thousands of optical coherence tomographs, which together formed thousands of images of the iris and pupil of a human eye. Optical coherence tomographs are a non-invasive scans routinely used in clinics to diagnose diabetic retinopathy together with macular degeneration. Optical coherence tomography is based on low-coherence interferometry, typically employing near-infrared light.
In total the researchers used an AI-based convolutional neural network to review more than 200,000 eye scans conducted, drawn from the optical coherence tomography images. The focus was on on two common causes of irreversible blindness: macular degeneration and diabetic macular edema. The accuracy achieved, by the end of the research, was 95 percent which was comparable to a well-trained ophthalmologist.
Further details of the research are outlined in the following video:
According to lead researcher Professor Kang Zhang: “Artificial intelligence (AI) has huge potential to revolutionize disease diagnosis and management by doing analyses and classifications involving immense amounts of data that are difficult for human experts — and doing them rapidly.”
The next stage of the research is to further improve the scanning for eye diseases and to begin using the artificial intelligence for diagnosing childhood pneumonia. The research has been published in the journal Cell, with the research paper headed “Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning.”
