May 29, 2019
Categorized: Healthcare | Research
Author: Bhavna Antony, IBM Research Scientist Australia
Many eye diseases that cause irreversible blindness are ones that develop slowly, showing little to no sign of vision threat until it is too late. Diabetic retinopathy and glaucoma are the leading and second leading cause of blindness worldwide, respectively. They currently affect 350 million individuals across the globe, with these numbers expected to rise due to the aging population and the increased occurrence of diabetes. These conditions, however, are treatable and vision loss can be prevented if the conditions are detected early.
Population screening is the answer, but an incomplete one. The expense associated with manual screenings done by a specialist is considerable. However, this doesn’t take into consideration the skill shortage and problems associated with accessibility to appropriate care in remote areas. An AI solution is ideal in this scenario as it could be deployed in urban as well as rural areas, and is a cost-effective solution for population screening.
Figure 1: Visualisation of network-detected regions in a glaucomatous (top row) and healthy (bottom row) eye.
As part of a team of scientists from IBM and New York University, my colleagues and I are looking at new ways AI could be used to help ophthalmologists and optometrists further utilise eye images, and potentially help to speed the process for detecting glaucoma in images. In a recent paper, we detail a new deep learning framework that detects glaucoma directly from raw optical coherence tomographic (OCT) imaging, a method which uses light waves to take cross-section pictures of the retina. This method achieved an accuracy rate of 94 per cent, without any additional segmentation or scrubbing of the data, which is usually time-consuming.
Currently, glaucoma is diagnosed using a variety of tests, such as intraocular pressure measurements and visual field tests, as well as fundus and OCT imaging. OCT provides an efficient way to visualise and quantify structures in the eye, namely the retinal nerve fibre layer (RNFL), which changes with the progression of the disease.
Beyond screening for common eye diseases, on-going research indicates that AI systems in the future will also be able to assess patients’ risks of cardiovascular complications (such as stroke) or even predict the development of neurological conditions such as dementia.