June 2023
Volume 64, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2023
Virtual monitoring of macular function with artificial intelligence in diabetic patients.
Author Affiliations & Notes
  • Laura Prieto Domínguez
    Ophthalmology, Hospital Clinico Universitario de Valladolid, Valladolid, Castilla y León, Spain
  • Alicia Canalejo
    Ophthalmology, Hospital Clinico Universitario de Valladolid, Valladolid, Castilla y León, Spain
  • María Isabel López Gálvez
    Ophthalmology, Hospital Clinico Universitario de Valladolid, Valladolid, Castilla y León, Spain
  • Footnotes
    Commercial Relationships   Laura Prieto Domínguez None; Alicia Canalejo None; María Isabel López Gálvez None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2023, Vol.64, 2645. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Laura Prieto Domínguez, Alicia Canalejo, María Isabel López Gálvez; Virtual monitoring of macular function with artificial intelligence in diabetic patients.. Invest. Ophthalmol. Vis. Sci. 2023;64(8):2645.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : In recent years, mobile applications that allow to detect macular disease have been developed. The aim of this observational prospective study is to analyze the possibility of using the OdySight® app in diabetic patients with and without diabetic macular edema (DME) and to study the frequency of use, the patient satisfaction degree with its use and the main app limitations that could condition its daily use.

Methods : A prospective, observational study was carried out in a sample of diabetic patients with and without DME who attended the hospital’s medical offices from February to March 2022. The far and near visual acuity (VA) of the patients was measured and they were then registered in the application, providing them with the necessary instructions to handle it. Subsequently, both the epidemiological data and the results obtained from the app were collected in a database including the follow-up time, how many times the test was performed, the VA values obtained in each test and the alerts related to decreases in VA. Finally, a questionnaire is given to all the participants following the model of the mHealth Satisfaction Questionnaire.

Results : The mean age of the 33 participants was 58.5 years with a standard deviation (SD) of 9.94. Of these patients, 64 eyes were studied, 33 of which had DME. No significant differences were found between the median VA measured by the application and the one previously obtained in the hospital (p=0.701). The patients have been in follow-up for an average of 45.8 days with a SD of 17,15 and 65.63% have performed the VA control test at least twice. During the follow-up period, alarms were triggered by 3 patients, 1 of them requiring immediate treatment. The satisfaction questionnaire was answered by 28 of the 33 patients, the 53,5% thought it was easy to perform the test and 82% of the participants recommend the application.

Conclusions : The OdySight® mobile app appears to reliably predict disease reactivations. However, its use in clinical practice has several limitations. The most important ones are the need of technological requirements, patients with advanced age and little familiarity with the use of new technologies. This is in contrast with the high degree of satisfaction found, so longer-term studies and with a larger sample of patients are needed to be able to define the current user profile.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

True alarm triggered in Odysight app by the appearance of DME.

True alarm triggered in Odysight app by the appearance of DME.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×