Consultation outcome

Automated grading in diabetic eye screening: rapid review and evidence map

From:
UK National Screening Committee
Published
18 May 2021
Last updated
25 November 2021 — See all updates
This consultation has concluded

Read the full outcome

Automated grading in the NHS Diabetic Eye Screening Programme cover sheet

PDF, 297 KB, 28 pages

This file may not be suitable for users of assistive technology.

Request an accessible format.
If you use assistive technology (such as a screen reader) and need a version of this document in a more accessible format, please email [email protected]. Please tell us what format you need. It will help us if you say what assistive technology you use.

Automated grading in the NHS Diabetic Eye Screening Programme review

PDF, 1.75 MB, 169 pages

This file may not be suitable for users of assistive technology.

Request an accessible format.
If you use assistive technology (such as a screen reader) and need a version of this document in a more accessible format, please email [email protected]. Please tell us what format you need. It will help us if you say what assistive technology you use.

Detail of outcome

The UK NSC received comments from 6 stakeholders.

Comments were reviewed by the UK NSC evidence team. They were discussed at the UK NSC meeting on 25 June 2021.

Background

Everyone with diabetes who is 12 years old or over is invited for diabetic eye screening once a year.

Currently in the NHS Diabetic Eye Screening (DES) Programme, health professionals take images of the back of each eye. Trained professionals working within the programme assess the images to determine if any eye disease is present. This process is known as ‘grading’.

Artificial intelligence (AI) systems are available that can be used within existing screening programmes to grade the images of eyes and potentially reduce the workload on the screening programme.

The UK National Screening Committee has not looked at the evidence on the use of AI in the DES programme before. Therefore, it is not currently recommended.

Recommendation

In 2021, the UK NSC looked at the use of automated grading in the DES programme.

The committee concluded that further research would be needed before it could be introduced. This is because, although there are AI systems available which are accurate enough to do the initial reading of images:

  • there is only limited evidence that it provides better health and value for money when compared to manual grading

  • there is uncertainty about the social and ethical aspects of using AI systems in the screening programme

The UK NSC is now working on a guidance document that outlines evidence requirements for the use of AI in the DES programme.

Original consultation

Summary

Consultation on the use of automated grading in the NHS Diabetic Eye Screening Programme.

This consultation ran from
to

Consultation description

The UK National Screening Committee (UK NSC) is inviting comments on the draft report on the use of automated grading in the NHS Diabetic Eye Screening Programme.

Please use the consultation response form below to submit your comments.

All submitted consultation comments, as well as UK NSC’s response, will be published at the same time as the final review.

Documents

Automated grading in the NHS Diabetic Eye Screening Programme

PDF, 1.6 MB, 176 pages

This file may not be suitable for users of assistive technology.

Request an accessible format.
If you use assistive technology (such as a screen reader) and need a version of this document in a more accessible format, please email [email protected]. Please tell us what format you need. It will help us if you say what assistive technology you use.

Updates to this page

Published 18 May 2021
Last updated 25 November 2021 + show all updates

  1. Automated grading in the NHS Diabetic Eye Screening Programme review document uploaded.

  2. The UK NSC has reviewed the evidence and provided a recommendation.

  3. Added consultation response.

  4. First published.

Sign up for emails or print this page

Related content