Insight Health Data Research Hub

19/03/2026

Using medical AI on ‘autopilot’ risks deskilling of clinicians, a team of doctors and aviation safety experts has cautioned. Instead, healthcare must embrace AI as a 'digital copilot'.

Writing for Nature Portfolio, the group of clinicians and flight safety specialists has set out five recommendations for the medical profession, informed by lessons from the aviation industry, which faced widespread loss of human skills after the adoption of autopilot.

The advice is co-authored by clinical researchers from Pearse Keane's research group at UCL Institute of Ophthalmology, Moorfields Eye Hospital, the INSIGHT Hub, who worked with the flight safety department of Lufthansa on the recommendations for future-proofing the clinical workforce and improving patient outcomes.

Read more:

As artificial intelligence (AI) becomes increasingly integrated into healthcare services, there are important lessons that the medical profession can learn from the aviation industry, which faced widespread loss of human skills after the adoption of autopilot.

16/03/2026

Discover the world-leading artificial medical intelligence work across UCL Institute of Ophthalmology and Moorfields Eye Hospital, supported by funding from Moorfields Eye Charity.

As Pearse Keane and Ariel Ong explain in the video below, development of healthcare AI is a complex process that begins with access to large-scale datasets and algorithm building, and encompasses the vital aspects of clinical evaluation, validation, safety and ensuring that AI algorithms work equally well on many different patient groups.

Pearse is Professor of Artificial Medical Intelligence at UCL, Consultant Ophthalmologist at Moorfields and Director of the INSIGHT Eye & Oculomics Health Data Research Hub. Ariel is a doctoral fellow, jointly funded by Moorfields Eye Charity and National Institute for Health and Care Research, and Data Lead for INSIGHT.

12/03/2026

Glaucoma is one of the leading causes of irreversible blindness worldwide, and detecting it in short-sighted (myopic) patients can be significantly more difficult.

As rates of both glaucoma and myopia rise, their combined influence on clinical services and patient outcomes remains an underexplored area of research - now being tackled by clinical optometrist and PhD researcher Caitlin Campbell.

Using large-scale anonymised health data through the INSIGHT Eye Hub at Moorfields Eye Hospital, Caitlin is investigating how myopia currently impacts glaucoma care, and how new clinical tests under development may be used to address the problem.

After completing her clinical training, including a pre-registration year at Moorfields, Caitlin began her PhD research into glaucoma and myopia at Ulster University, working in collaboration with researchers at Moorfields, UCL Institute of Ophthalmology and Cardiff University, under the supervision of Pádraig Mulholland.

Find out more about Caitlin’s research using anonymised NHS health data. https://www.insight.hdrhub.org/post/researchers-use-health-data-to-tackle-a-growing-challenge-in-glaucoma-care

26/02/2026

The first fundus photographs of the human eye were reproduced 140 years ago by William Thomas Jackman and JD Webster in an 1886 London edition of "The Photographic News". The images were subsequently published in international journals of the day, along with the technique, which entailed an exposure time of 2 minutes, 30 seconds.

Since then, Colour Fundus Photographs (CFPs) have become the dominant eye imaging modality for assessing retinal conditions, especially diabetic retinopathy. Images are typically captured in a fraction of a second.

At Moorfields Eye Hospital, around 50,000 images each month are added to INSIGHT's data repository and made ready for medical research. As our data visualisation shows, a significant proportion of these are CFPs, alongside Optical Coherence Tomography (OCTs), Scanning Laser Ophthalmoscopy (SLO), Fluorescein Angiography (FA), and Fundus Autofluorescence (AF). Together they comprise the world's largest resource of ophthalmic imaging linked to medical records.

Find out more about the imaging modalities represented in the INSIGHT data resource and the conditions they cover: https://lnkd.in/e7X5KQcK

Historical image courtesy Internet Archive

10/02/2026

In the latest edition of Acuity, the professional development journal of the College of Optometrists, journalist Helen Bird explores how AI is transforming eye care by supporting earlier disease detection, clearer tracking and more accurate referrals.

The feature is part of a series on and draws on expertise shared by Pearse Keane and Anthony Khawaja of UCL Institute of Ophthalmology and Moorfields Eye Hospital.

Reproduced with kind permission of Acuity's editors.

Decoding Disease: the latest edition of Acuity, the journal of the College of Optometrists, explores how AI is transforming eye care by supporting earlier disease detection, clearer tracking and more accurate referrals.

08/02/2026

For the latest episode of Mork Unfiltered, Patrick Mork invites Pearse Keane to share how he built a career in and helped to forge the field of artificial medical intelligence, driven by the need to prevent people from going blind.

Together they cover:
• The "Sliding Doors" Moment: How a LinkedIn message to Google DeepMind influenced the course of medical history.

• The Power of : Why the eye is a "window to the rest of the body" and can predict the risk of a heart attack.

• Forged in the Trenches: Surviving the "Special Forces" training of a surgical internship and what it teaches you about human capacity.

• & Open Source: The decision to share the world's first AI foundation model for eye health with the public to accelerate patient benefit at scale.

• Defining "Artificial Medical Intelligence": Why we need a new kind of leadership to figure out what AGI looks like when applied to healthcare.

Pearse is professor of artificial medical intelligence at UCL Institute of Ophthalmology, consultant ophthalmologist at Moorfields Eye Hospital and director of INSIGHT Eye Hub.

Watch the video: https://youtu.be/ppuGE7_Nzp8

Listen on Spotify: https://open.spotify.com/episode/0qUNxAFZPeeYzV2PxLbfGQ?si=d5ca0b4c2dfc4b38

Listen on Apple Podcast: https://podcasts.apple.com/us/podcast/how-a-single-eye-scan-could-save-your-life/id1824866742?i=1000748382075

hashtag hashtag hashtag

👉 Download Mork’s Guide to Resilience: https://aa5c-ea.systeme.io/resilience🌟 Hire Patrick to speak about Purpose, Resilience, and Leadership: https://patr...

02/02/2026

Healthcare AI improves diagnostics and patient outcomes in places where resources are limited, finds new three new studies led by Duke-NUS Medical School with collaborating researchers including colleagues at UCL Institute of Ophthalmology Moorfields Eye Hospital and INSIGHT Eye Hub.

Click below for the full story ⬇️

Did you know that AI tools can guide doctors in making critical decisions and improve healthcare access, especially in resource-limited settings?

In one example, Assoc Prof Liu Nan and his team, along with their collaborators, demonstrated that existing AI models for predicting patient outcomes after cardiac arrest could be adapted for data-scarce regions via transfer learning. By removing the need to train models from scratch, this technique reduces costs and development time for introducing healthcare AI to new environments.

The team also called for the formation of an international consortium to help healthcare systems around the world leverage AI’s strengths safely and responsibly.

Read more about it from the link in the comments.

28/01/2026

Data Protection Day, marked annually on 28th January, commemorates the Council of Europe’s Convention 108, introduced in 1981 as the first legally binding treaty protecting privacy in the digital age.

45 years on, the day is marked globally to raise awareness of how personal data is collected, stored and used. In the context of healthcare data, this may be more important than ever, with the rise in wearables, remote monitoring tools and digital health apps accelerating the amount of personal health data being collected.

As the world’s largest resource of eye imaging data linked to clinical records, collected through routine appointments at Moorfields Eye Hospital, INSIGHT meets comprehensive governance and ethical standards for enabling and supporting research and innovation into eye health and systemic health.

"Data protection isn't a barrier to medical progress – it's the foundation that makes progress possible," says INSIGHT Director Pearse Keane. "Without public trust, built on transparent and rigorous safeguards, research simply cannot happen."

Click on the image below to read more about data protection at INSIGHT, and how patients play a central role.

INSIGHT enables research across Moorfields, UCL Institute of Ophthalmology and the wider healthcare and medical ecosystem. Clinical researchers and data scientists working with us are supported by NIHR Moorfields Biomedical Research Centre, Moorfields Eye Charity Friends of Moorfields

Marked annually on 28th January, Data Protection Day is a timely opportunity to highlight how INSIGHT at Moorfields safeguards data, while enabling world-leading medical research.

19/01/2026

In conversation with The Ophthalmologist, our colleagues Pearse Keane and Jeffry Hogg from Moorfields Eye Hospital discuss the role of in improving the management of patients with Age-related Macular Degeneration

AI isn’t here to replace ophthalmologists - it’s here to strengthen clinical decision-making.

We spoke with Jeffry Hogg and Pearse Keane (Moorfields Eye Hospital, London) about how AI-enabled OCT analysis could help close treatment gaps in neovascular AMD, reducing both under- and overtreatment in real-world care.

“When ophthalmologists work together with AI, they rank the correct diagnosis higher and enrich their management plans.”

🔗 Read the full interview: https://ow.ly/t69H50XTrfE

12/01/2026

A commonly used eye injection can effectively manage a previously untreatable condition, hypotony, reports a new paper published in British Journal of Ophthalmology.

Hypotony is characteristed by abnormally low intraocular pressure, which can cause the eye's internal structures to distort, leading to vision loss. It affects around 100 people in the UK each year. The standard treatment has been to fill the eye with silicon gel. However, this can have toxic side effects.

Now, research led by colleagues at Moorfields Eye Hospital, at the world's first dedicated hypotony clinic, has found that ocular injections containing HPMC (hydroxypropyl methylcellulose) - a low-cost gel used in most eye surgery - can be safely used to increase the size and pressure of eyes.

As reported by the BBC: https://www.bbc.co.uk/news/articles/c89qyv98lzdo

Read the paper: https://bjo.bmj.com/content/early/2026/01/04/bjo-2025-327866?rss=1

Nicki's eye had collapsed in on itself, but a new gel injection method has saved her vision.

29/12/2025

Over the past year, INSIGHT marked some significant milestones, and enabled world-leading research in medical artificial intelligence - laying the foundations for 2026 and beyond. Here are a few highlights:

- A collaboration with US biotech insitro to build a foundation model for drug discovery in and neurology; with researchers from the Pearse Keane medical AI group that spans UCL Institute of Ophthalmology, Moorfields Eye Hospital NIHR Moorfields Biomedical Research Centre co-leading development for INSIGHT. The project was championed by patient advocate Elaine Manna, who sadly passed away in July. Elaine's legacy lives on through this and other work that will benefit future generations.
https://www.insight.hdrhub.org/post/insitro-and-uk-s-insight-at-moorfields-eye-hospital-announce-collaboration-to-expand-research-effort

- An AI tool that can predict eye damage from the autoimmune drug hydroxychloroquine years before signs appear. https://www.insight.hdrhub.org/post/ai-predicts-eye-damage-from-autoimmune-drug-years-before-symptoms-appear

- Launch of Global RETFound, a consortium of research groups from more than 100 countries building the first AI model in healthcare with globally representative data. https://www.insight.hdrhub.org/post/quest-to-build-the-first-medical-ai-foundation-model-with-globally-representative-data

- Work led by Shafi Balal and colleagues, predicting when young people with the sight-threatening eye condition kerataconus need treatment to stabilise the cornea and prevent loss of vision. https://www.insight.hdrhub.org/post/ai-detects-eye-changes-that-lead-to-sight-loss-in-young-people

- An AI system that can accurately assess disease activity in wet AMD, potentially transforming how the NHS monitors and manages the condition. hhttps://www.insight.hdrhub.org/post/ai-can-identify-treatment-gaps-for-patients-with-macular-disease-study-shows

- Starting in 2026, a new initiative supported by UK Research and Innovation Medical Research Council and NIHR to scale INSIGHT and create a national resource for eye health and research.
https://www.insight.hdrhub.org/post/new-initiative-to-transform-uk-eye-health-research-by-linking-national-data-resources

This progress is made possible thanks to the thousands of patients who attend routine appointments at Moorfields every month, contributing to INSIGHT's anonymised data pipeline of more than 30 million images linked to clinical records - the world's largest ophthalmic bio-resource for healthcare research.

As always, Moorfields Eye Charity and Friends of Moorfields have proved invaluable allies in our work.

19/12/2025

Share your experiences and views on AI in healthcare with the Medicines & Healthcare products Regulatory Agency and help to shape the standards for new AI technologies in the NHS and wider health sector. The agency has launched a call for evidence, open to everyone. Closes 2nd Feb 2026.