We are searching data for your request:
Upon completion, a link will appear to access the found materials.
App can detect cancer in photos
A free smartphone app can detect a rare form of eye cancer by evaluating captured images. This cancer usually affects babies and children under the age of five. If the disease is diagnosed early, there is a good chance of a cure.
In the current study by Baylor University in Waco, Texas, an app was tested that can diagnose a rare form of cancer in the eyes. The results of the study were published in the English-language journal "Science Advances".
App detects white gloss in the pupil on photos
The app looks for an abnormal white luster in the pupil, which can be recognized in photos and during eye examinations. Such a white shine can be a symptom of retinoblastoma, a form of cancer that mostly affects babies and children under the age of five. If the cancer is recognized early enough, it can be treated.
More than 52,000 images were evaluated
The app has been tested on more than 52,000 photos of forty children. Twenty children suffered from various eye diseases, and there were also twenty healthy controls. In four out of five children with eye diseases, the app recognized the white gloss in the eye from photos taken an average of 1.3 years before the diagnosis. The new technology will hopefully enable young people to be diagnosed and treated early, which will greatly increase their chances of being cured, the research team said.
Technology evaluates images on your own smartphone
The new technology CRADLE is trained to automatically analyze photos in the gallery of a smartphone. The CRADLE application privately analyzes digital images that are stored directly on the user's device without the need to upload a photo. The study suggests that a parent can use a smartphone application like CRADLE to detect leukocorie (white gloss) early on. The app has also been tested in children with Coats' disease, which causes abnormal blood vessels in the retina and can cause blindness.
Accuracy of the app was approximately 80 percent
In 80 percent of children with eye diseases, the app correctly identified a leukocorie in the photos. So far, a so-called red reflex test has been used to diagnose the disease. This is a non-invasive method to look for early warning signs of serious eye diseases. When the eye is healthy, a camera flash usually illuminates the blood-rich retina. However, this requires that both eyes look directly at the camera, which makes it difficult to assess casual shots. Digital photography, including personal photos taken during child-friendly activities and in child-friendly places, is a practical addition to conventional leukocorie screening. Any photo taken by a parent throughout childhood has the potential to be a test for To represent leucocorie.
The reliability of the app increases by evaluating multiple images
In the future, the app is to be further improved to produce fewer misdiagnoses. The app is more reliable when it comes to evaluating large amounts of photos, the team reports. The app can currently be downloaded, the prototype was immediately uploaded to the Apple App Store and on Google Play under the name White Eye Detector. The American Academy of Pediatrics emphasizes that all children with signs of white shine should be examined by an ophthalmologist. (as)
Author and source information
This text corresponds to the specifications of the medical literature, medical guidelines and current studies and has been checked by medical doctors.
- Micheal C. Munson, Devon L. Plewman, Katelyn M. Baumer, Ryan Henning, Collin T. Zahler et al .: Autonomous early detection of eye disease in childhood photographs, in Science Advances (query: 04.10.2019), Science Advances