Cradle app can detect eye diseases in children

4 October 2019 (Last Updated October 4th, 2019 10:42)

Baylor University researchers have developed and studied a prototype smartphone app, Cradle, for the detection of early signs of various eye disorders, including an aggressive eye cancer called retinoblastoma, in children.

Cradle app can detect eye diseases in children
The technology can help parents screen for common eye diseases in children. Credit: virginia332 from Pixabay.

Baylor University researchers have developed and studied a prototype smartphone app, Cradle, for the detection of early signs of various eye disorders, including an aggressive eye cancer called retinoblastoma, in children.

Study findings indicate that the Cradle app can effectively help in clinical screenings for leukocoria, a primary symptom of retinoblastoma.

The app detects leukocoria by looking through family photographs for traces of any abnormal reflections from the retina. Also, it can help parents screen for other common eye diseases.

During the study, Baylor University research team evaluated the sensitivity, specificity and accuracy of the prototype app via examination of more than 50,000 photographs taken before diagnosis.

The app identified leukocoria in 80% of the children diagnosed with eye diseases. Cradle was able to detect the condition in photographs taken on an average of 1.3 years before the official diagnosis.

Conventional screening approaches detect leukocoria in only 8% of cases, said the researchers.

The sensitivity of the app was found to be more than 80% in children aged two years and below. The team added that the ability of Cradle’s algorithm to identify even slight instances of leukocoria has improved.

Baylor University chemistry and biochemistry associate professor Bryan Shaw said: “We suspected that the app would detect leukocoria associated with other more common disorders and some rare ones.

“We were right. So far parents and some doctors have used it to detect cataract, myelin retinal nerve fibre layer, refractive error, Coats’ disease, and of course retinoblastoma.”

The Science Advances journal has published the study findings.

Currently, researchers are using the app to analyse nearly 100,000 photos to identify additional features that could help reduce false positives.