July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Cost-Effectiveness Analysis of an Artificial Intelligence-Assisted Deep Learning System Implemented in the National Tele-Medicine Diabetic Retinopathy Screening in Singapore
Author Affiliations & Notes
  • Yuchen Xie
    Singapore Eye Research Institute, Singapore, Singapore
  • Quang Nguyen
    Singapore Eye Research Institute, Singapore, Singapore
  • Valentina Bellemo
    Singapore Eye Research Institute, Singapore, Singapore
  • Michelle Y.T. Yip
    Duke-NUS Medical School, Singapore
  • Xin Qi Lee
    Singapore Eye Research Institute, Singapore, Singapore
  • Haslina Hamzah
    Singapore National Eye Centre, Singapore
  • Gilbert Lim
    National University of Singapore, Singapore
  • Wynne Hsu
    National University of Singapore, Singapore
  • Mong Li Lee
    National University of Singapore, Singapore
  • Jie Jin Wang
    Duke-NUS Medical School, Singapore
  • Ching-Yu Cheng
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Eric Andrew Finkelstein
    Duke-NUS Medical School, Singapore
  • Ecosse Luc Lamoureux
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Gavin Siew Wei Tan
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Tien Yin Wong
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Daniel SW Ting
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore
  • Footnotes
    Commercial Relationships   Yuchen Xie, None; Quang Nguyen, None; Valentina Bellemo, None; Michelle Yip, None; Xin Qi Lee, None; Haslina Hamzah, None; Gilbert Lim, EyRIS (P); Wynne Hsu, EyRIS (P); Mong Li Lee, EyRIS (P); Jie Jin Wang, None; Ching-Yu Cheng, None; Eric Finkelstein, None; Ecosse Lamoureux, None; Gavin Tan, None; Tien Wong, EyRIS (P); Daniel Ting, EyRIS (P)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 5471. doi:https://doi.org/
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yuchen Xie, Quang Nguyen, Valentina Bellemo, Michelle Y.T. Yip, Xin Qi Lee, Haslina Hamzah, Gilbert Lim, Wynne Hsu, Mong Li Lee, Jie Jin Wang, Ching-Yu Cheng, Eric Andrew Finkelstein, Ecosse Luc Lamoureux, Gavin Siew Wei Tan, Tien Yin Wong, Daniel SW Ting; Cost-Effectiveness Analysis of an Artificial Intelligence-Assisted Deep Learning System Implemented in the National Tele-Medicine Diabetic Retinopathy Screening in Singapore. Invest. Ophthalmol. Vis. Sci. 2019;60(9):5471. doi: https://doi.org/.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Deep learning systems (DLS) showed clinically acceptable performance in the detection of diabetic retinopathy (DR) using retinal images, although the cost-effectiveness remains unknown. This study aims to examine the cost-effectiveness of an innovative AI-assisted DLS screening model (Model A), compared to the current manual grading system (Model B) in Singapore Integrated Diabetic Retinopathy Screening Program (SiDRP) (Fig 1).

Methods : A hybrid decision tree/Markov model was built to simulate the DR progression for a hypothetical diabetic cohort. Patients enter the model at age 46 years (mean age onset of Diabetes in Singapore), never screened for DR. Decision tree components of the model captured entire progress from primary care screening to specialist consultation and treatment. Model parameters included direct medical costs, utility values of health states, DR transition probabilities and the screening performance. All costs consisted of DR screening, clinic visits and follow-up treatments. Values for all transition probabilities and utility values of DR, including no DR, mild DR, moderate DR, vision-threatening DR, blindness, and death were derived from relevant studies in the literature. The prevalence of DR and screening performance (sensitivity and specificity) of Models A and B, had been reported by Ting et al. (JAMA 2017). The primary outcome measures were the total cost incurred by the healthcare system, and the total quality-adjusted life-years (QALYs) gained per patient. One way and probabilistic sensitivity analyses were performed.

Results : Over a lifetime, a patient with DR would incur a total cost of S$1177 and S$1312 under Models A and B (Table 1), respectively. From the health system perspective, Model A results in a lifetime cost-saving of S$135 per patient while maintaining comparable QALYs gained. Given that the projected screening visits with diabetes will rise up to approximately 250,000 patients by 2025, the present value of future cost savings associated with AI-assisted DLS screening is estimated to be $33.8 million.

Conclusions : While maintaining a comparable clinical outcome, AI-assisted DLS screening model is less costly than the current manual grading system, demonstrating the strong economic rationale and need for AI-integrated DR screening program for Singapore and elsewhere.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

 

×
×

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.

×