J Korean Ophthalmol Soc.  2020 Jun;61(6):624-629. 10.3341/jkos.2020.61.6.624.

Quantification of Vascular Tortuosity by Analyzing Smartphone Fundus Photographs in Patients with Retinopathy of Prematurity

Affiliations
  • 1Department of Ophthalmology, Dong-A University College of Medicine, Busan, Korea
  • 2Department of Electronics Engineering, Dong-A University College of Engineering, Busan, Korea

Abstract

Purpose
To determine the usefulness of measuring the tortuosity of retinal arteries using smartphone fundus photographs to quantify plus disease in retinopathy of prematurity (ROP) patients.
Methods
Digital fundus photographs were taken with a smartphone of 116 eyes involving 58 premature infants. The tortuosity of retinal arteries named as the tortuosity index (TI) was measured with Image J software. Patients were classified into the treated and control groups and the TIs of the two groups were compared. A receiver operating characteristic curve was constructed and areas under the curve (AUC) were calculated to quantify the diagnostic utility of TI.
Results
Fundus photographs of 98 eyes of 50 premature infants were analyzed; 38 eyes of 20 infants were the treated group and 60 eyes of 30 infants were the control group. The TI was 1.231 ± 0.156 in the treated group before treatment, which was significantly larger than 1.062 ± 0.019 in the control group. After treatment of the treated group, the TI was 1.100 ± 0.093, which was significantly smaller than before treatment. The AUC was 0.926, and using a cut-off value of 1.095, the sensitivity and specificity of TIs for requiring treatment were 85.1% and 95.5%, respectively.
Conclusions
The tortuosity of retinal arteries was measured by analyzing smartphone fundus photographs to quantify plus disease in ROP patients. This method may be helpful for screening, follow-up, and treatment decisions for ROP patients.

Keyword

Plus disease; Retinopathy of prematurity; Smartphone; Vascular tortuosity
Full Text Links
  • JKOS
Actions
Cited
CITED
export Copy
Close
Share
  • Twitter
  • Facebook
Similar articles
    DB Error: unknown error