# Implementing an artificial intelligence system into a diabetic eye screening programme in Tanzania

## TL;DR

This paper discusses implementing an AI system to improve diabetic retinopathy screening in Tanzania, where specialist eye care staff are scarce.

## Contribution

The novel contribution is the practical implementation of AI in a real-world diabetic eye screening program in a low-resource setting.

## Key findings

- AI systems can help task-shift diabetic retinopathy screening away from specialist staff.
- Implementation of AI in northern Tanzania's screening program is described.
- AI has potential to improve eye care in regions with limited specialist resources.

## Abstract

Tanzania has the highest age-adjusted prevalence of diabetes in sub-Saharan Africa. Diabetic retinopathy, a common complication, is a significant cause of vision loss; but with effective screening and treatment this often can be prevented. However, with very few specialist eye care staff in Tanzania this is a major challenge. Artificial intelligence (AI) systems, which automate clinical decision making and therefore task-shift away from specialist staff, could contribute to improved diabetic retinopathy screening services in low-resource settings. This article describes our experiences of selecting, procuring and implementing an AI system into a regional diabetic eye screening programme in northern Tanzania.