Routine eye scan could predict 10-year heart attack and stroke risk
Researchers used artificial intelligence (AI) to analyse digital retinal photographs, which are taken to capture images from the back of the eye to look for certain conditions.
The technology was able to produce personalised risk predictors in less than a second.
It is hoped the simple test could one day be offered alongside blood pressure and cholesterol checks to allow doctors to treat patients earlier.
For the study, published in the journal Cardiovascular Diabetology, researchers at the University of Dundee trialled the AI software on the eye scans of people with type 2 diabetes.
Those with the condition have routine eye tests to check for diabetic retinopathy, a condition where high blood sugar levels damage the blood sugar in the retina.
Dr Ify Mordi, British Heart Foundation research fellow at the University of Dundee and consultant cardiologist, said: 'It may be surprising, but the eyes are a window to the heart.
'If there is damage or narrowing of the blood vessels at the back of the eye, there is a good chance that will also be seen in the blood vessels further inside the body, supplying the heart, which could lead to a heart attack or stroke.'
Experts instructed the technology to initially look for warning signs on the images, such as blood vessel narrowing or blockages.
It was then permitted to look for any detail in the photographs, such as the size or arrangement of blood vessels, before being trained using 4,200 images.
The AI tool was able to predict the people who would go on to have 'major cardiovascular events', like heart attacks or strokes, within a decade with 70% accuracy.
Dr Mordi added: 'This is a one-stop scan which is routinely performed and takes less than a minute.
'It could be an important part of the package, alongside blood pressure and cholesterol checks, in identifying people who could benefit from medication or lifestyle changes.'
Professor Bryan Williams, chief scientific and medical officer at the British Heart Foundation, said: 'The more accurately we can detect someone's risk of a heart attack or stroke, the better the opportunities to prevent these happening.
'Cutting-edge innovations, like the use of retinal scans alongside health checks, could play a role in improving risk prediction, which is important if we are to reach the British Heart Foundation's goal to prevent 125,000 heart attacks and strokes in the UK by 2035.
'However, more research is needed to show this prediction accuracy is robust, and to determine the feasibility of incorporating retinal scans into clinical practice.'
Reacting to the study, Dr Clare Jonas, research lead at the Stroke Association, said: 'AI analyses of health data, gathered through screening technology such as this eye scan, have exciting potential for better early detection, especially for those with diabetes who have a greater chance of having a stroke.
'Advances like these could make it easier for people at risk of stroke to take the right steps to stop it from happening.
'Whilst there are still many practical questions to answer about whether, when and how AI could be used in clinical practice, we hope the Government's 10 Year Health Plan will take such technological advances into account to stop more strokes from devastating people's lives.'
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
2 minutes ago
- Forbes
JetBrains Aims Boosters At No-Code Web Development
Build a website, or web app. Simple, right? Although not many of us do. There are plenty of tools out there such as Wix, Squarespace, Weebly and of course WordPress itself. Then there are the GoDaddy-type, Duda and Jimdo services that aim to simplify as much as possible, but at the end of the day, they still present a technology proposition that many average users consider to be out of reach. Each website building service has its own variety of themes and level of customizability, each has its own pricing structure and onward licensing considerations, each has different levels of 'robustness' (always important if a website or web app is intended to host e-commercce services)… and each has its own approach to more granular (but still important) considerations such as multi-lingual human language support. Optimization & Formalization In our web-evolved world of search engine optimization and formalized website indexing, the majority of brands and services on the web have been professionally presented. This makes any amateur approach to website or web app development today (something that we might have considered as hobbyist and cute back in the 1990s) now have a potentially negative impact on public perception. As UK-based bespoke web design agency TeeGee Digital puts it, 'With a free site builder, Google will not be able to verify your website using webmaster tools. All of the leading search engines use complicated search algorithms to rank and compare websites. These can be very complicated to someone with little web design experience however, for a professional these insights can be very helpful in monitoring and improving your website.' Integrated development environment and developer project platform company JetBrains thinks it has a way forward for users who want a no-code option for web development. Although the company is known for its deep tech, including the IntelliJ IDEA developer IDE and its Junie AI coding agent for software engineers, the company also produces Kineto, technology built to create, deploy and maintain ready-to-use websites and applications. Beyond The Vibe Coding Groove 'Delivering even a simple to-do [web] app requires a hefty technology stack, a team of professionals and hours of testing and maintenance,' noted JetBrains product leader Andrew Zakonov and software engineering development team leader Vladimir Krasnotsvetov in a recent technical document. 'Website constructors increase web development speeds, but they still require user input to actually build a website… and while there are generative AI-generative platforms, these still need users to do some (vibe) coding.' The pair claims Kineto is capable of building bridges between creative people, small businesses and no-code platforms, contributing to the emergence of a creative developer economy where people will create their own personal web applications. 'The key principle behind Kineto is no code for [users] at all. Our tools take care of the coding, so [users] can focus on being creative,' note Zakonov and Krasnotsvetov. What is A Single-Value Web App? JetBrains says that Kineto works well for creating what it calls small 'single value' applications. So that's unlikely to be the next NASDAQ portal or the next Netflix, this would more likely be useful for building your own custom-aligned fitness tracker app, for creating a blog, or some kind of personalized online planning tool. That single value app is presented with a user interface, an appropriate level of backend infrastructure provisioning (to handle computations, input/output of data, external API connects and so on) and a database service. All of which can be managed through a chat-based interface that works in natural language. Kineto takes about 20 minutes to build its first prototype once instructed, the user can then work out how they want to refine, extend and improve it. 'The first step with Kineto is to formulate the prompt with a detailed explanation of what your app should do and what it should look like. Every prompt is processed by Kineto and translated into appropriate functionality. You don't need to explain every single button or provide extensive references; Kineto will think of all that for you and then propose a suitable interface and design,' said Zakonov and Krasnotsvetov. 'It's okay if you don't know exactly what you want your app to look like. Kineto will ask you to choose a basic design template, color scheme and font in order to give it a solid starting point from which to build your app.' Currently going through its early access period, JetBrains says aims to make this technology as mainstream as it can in the weeks and months ahead. Competive Analysis: No-code Web Development While this technology is still somewhat embryonic, it's tough to know how kind to be in terms of its potential, or how critical to be to its wider approach. Over and above the competing technologies mentioned at the start, it's worth remembering that smaller brands such as Adalo offers Apple iOS and Android mobile app development with a drag-and-drop interface. Another example, the so-called 'decision canvas' offered by Sitecore Personalize is a tool that offers both low-code and no-code option. More aligned towards omnichannel customer data tools used in the appropriate business context (it's an enterprise tool, basically), this software still reaches downward (no disrepect intened) so non-technical business users can create intricate personalization paths for customers. Bubble is also known for drag-and-drop development; the company is known as the 'full stack no-code app builder' and says it's the AI app development platform for 'founders and businesses' building serious, scalable apps through visual development for web and native mobile, all with no code required. Then there's Parabola for no-code data processing and automation (okay, more for workflows than a full website), plus Tilda for no-code website builds with design tools. This is a growing market. Any one of those brands could be the next Twitter for AI web development, or we might see a major brand like SAP Build Apps (formerly SAP AppGyver) democratize itself and become more all-user accessible if the German softwarehaus sees fit to widen its roadmap. This development will be interesting for some, but it may carry an air of 'citizen development' for others. While there's no point in fighting progress and the rise of citizen coders, citizen data analysts, citizen web developers and citizen X (i.e. everything, via low-code no-code and artificial intelligence) will continue to burgeon, there is clearly a good amount of AI working here. That presence of AI services in place at the backend is warm and fuzzy at the outset; after all, this is kind of microwave age convenience for web apps (just 20 minutes and your app will be ready, caution: filling is hot) so what's not to like? Well, users should arguably be wise to think about AI-driven backend platforms in terms of their ability to deliver robust outputs without hallucinations. Humans in this particular loop appear to still be very much needed.
Yahoo
30 minutes ago
- Yahoo
The best (and worst) times to eat dinner, according to an expert
UK restaurants have seen a surge in 6pm bookings. Gone are the days of 9pm restaurant bookings and late-night Mediterranean-style banquets. For now, it's all about the 6pm dinner. According to new data from online reservation service OpenTable, the number of 6pm bookings in the UK have seen a sharp rise of 11% year-on-year, while hospitality tech company Zonal adds that 6.12pm has become the new national average. Interestingly, 5pm slots have also risen by 10% in the past year. Responding to the trend, which The Times has attributed to Gen Z, restaurants across London have begun introducing special menus, specifically catered towards early diners – much like pre-theatre offers. So, what's behind this trend? And is there a "right" time to eat your tea? Why are people eating earlier? There are a number of possible contributing factors – not least the influence of social media and the rise in hybrid working models post-COVID. Hannah Belsham, health and wellbeing physiologist at Nuffield Health's Highgate Hospital, tells Yahoo UK: "With more people working from home or having flexible hours, it's potentially easier to sit down to dinner sooner, with less commuting and more time to get extra commitments done sooner in the day." Even those commuting to the office may wish to dine earlier if they're eating out, especially if they need to catch a train. The rise in "What I Eat In a Day" videos may also be influencing this trend, along with Gen Z's move away from drinking culture. It's less about pints at the pub after work and more about wellbeing and nutrition. Earlier dinner times have long been favoured among Brits, though, with a 2020 YouGov poll revealing most UK adults (34%) sit down to eat between 6-6.59pm. The same poll showed 23% eat between 7pm-7.59pm and 25% between 5pm-5.59pm. Parenthood likely also contributes to this trend, with many adults opting to eat at the same time as their kids for ease. That said, research by Sainsbury's from January 2021 revealed that just 28% of households were sharing the same meal in the evening and 55% stated they struggled to sit down together for family meals. The impact of late-night eating According to Belsham, eating a large meal late at night can interrupt our circadian rhythm – the body's natural internal clock which regulates sleep, hormone production, and metabolism. "Digestion increases metabolic activity, raising body temperature and diverting blood flow to the gut when the body is preparing for rest. This misalignment can impair sleep onset and quality," she explains. Moreover, late-night eating is linked to impaired glucose metabolism. "Studies show that consuming high-calorie meals closer to bedtime can reduce insulin sensitivity, leading to higher blood sugar levels overnight, which over time may increase risk for metabolic conditions such as type 2 diabetes." But then, when we look to our Mediterranean counterparts, who generally eat far later than we do, they are the picture of health – with one study from Harvard suggesting they live longer, have better brain and heart health and have reduced risk of cancers. In fact, the Mediterranean diet and lifestyle is something many of us Brits try to emulate, owing to its many benefits. On this topic, Belsham notes: "Despite later dinner times, Mediterranean populations often exhibit excellent metabolic health and lower rates of chronic disease. This seeming paradox can be explained by meal composition and lifestyle. Mediterranean dinners emphasise nutrient-rich foods like vegetables, legumes, whole grains, lean proteins, and healthy fats, foods that produce a more gradual rise in blood sugar and are easier to digest. "Additionally, portion sizes tend to be moderate, and meals are typically social and relaxed events, which may reduce stress-related digestive issues and improve nutrient absorption. The physical activity common in these cultures also helps regulate metabolism and supports cardiovascular health," she adds. The best (and worst) times to eat your dinner As Belsham says, individual factors such as your lifestyle, digestive health, and sleep patterns will influence the optimal timing of your heaviest meal. "If you experience acid reflux, poor sleep, or discomfort after late dinners, it's worth experimenting with eating earlier (ideally finishing at least two to three hours before bedtime)," she recommends. On the other hand, if you routinely eat later without any negative symptoms and maintain good sleep and energy levels, your current routine may be well suited to your biology, per the expert. Practical recommendations Aim to complete your largest meal two to three hours before you go to bed to allow digestion to progress before sleep. Prioritise balanced meals rich in fibre, healthy fats, and lean proteins to stabilise blood sugar and reduce digestive strain. If you prefer later dinners, keep portions moderate and avoid overly fatty or spicy foods that can interfere with rest. Monitor your body's responses and adjust timing accordingly for optimal wellbeing. Ultimately, Belsham says, while there is no universal "perfect" dinner time, understanding how meal timing interacts with your circadian biology and lifestyle can help you make informed choices that support digestion, sleep, and long-term health. Read more about health and nutrition What you need to know about Japanese walking and how to do it every day (Yahoo Life UK, 4-min read) How to BBQ – and not poison your guests (Yahoo Life UK, 7-min read) Why you shouldn't drink iced coffee on a hot day (Yahoo Life UK, 3-min read) Solve the daily Crossword
Forbes
an hour ago
- Forbes
Why CMOs Should Stop Chasing ROAS And Start Measuring Impact
Liam Wade - Performance Director at Performance Marketing Agency, Impression. In a competitive landscape, the way you measure performance shapes the way you invest—and that can make or break your growth. If your decisions are based on surface-level metrics like return on ad spend (ROAS), while your competitors are optimizing against true business impact, you're not just at a disadvantage. You're playing a different game entirely. ROAS has become the go-to metric for many. It's quick, accessible and easy to drop into a slide deck for the boardroom. A high ROAS can boost confidence, justify more budget and make teams feel as if they're winning. However, there's a problem: ROAS rarely tells you what's actually working. It shows correlation, not causation, and that distinction matters. The companies that win in today's landscape aren't just running ads. They're measuring what those ads really did—not what they touched, but what they changed. The Problem With ROAS ROAS is often treated like a bottom-line metric. Spend $1, get $5 back? Sounds good? Here's the catch: It tells you what happened after someone saw or clicked an ad. It doesn't tell you if the ad had anything to do with the outcome. This becomes a problem when the campaigns showing the highest ROAS, like branded search or remarketing, are often reaching people who were already likely to convert. These campaigns look brilliant, but they're not creating new demand. They're just scooping up what was already there. To make things worse, ROAS is built on tracking models that rely heavily on cookies and click paths. In theory, they map a clean journey. In practice, they miss most of what's going on—privacy restrictions, ad blockers, dark social and offline activity. You're seeing just a fraction of the picture and assuming it's the whole. When the same platform runs your ads and tells you how they performed, let's just say they don't have much incentive to be neutral about it. The Cost Of Optimization Obsession Over time, this obsession with ROAS creates an even bigger issue. Teams stop asking the big questions. Instead, they focus on squeezing marginal gains out of short-term tactics. You've probably seen this play out. Endless A/B tests. Bid adjustments. Hyper-specific audience filters. It looks like progress, but it can actually distract from more important questions: Are we bringing in new customers? Are we building a brand people care about? Worse, chasing ROAS can quietly damage long-term performance. Eventually, the algorithm starts chasing easy wins (existing customers, bottom-funnel conversions), while upper-funnel or experimental campaigns quietly get deprioritized—not because they don't work but because they don't show up in the right column of a dashboard. It's a pattern I've seen too often, what I call dashboard delusion. Everything looks great, but it's actually at the expense of real growth. What CMOs Should Measure Instead Here's the question I always come back to: What would have happened if we hadn't run this campaign? That's the heart of incrementality. It doesn't ask what an ad touched. It asks what it caused. That's the only way to know whether your marketing is really moving the needle or just riding the wave. There are a few ways to approach it: • Run geo tests. Show ads in one region, hold them back in another, and compare. • Use holdout groups. Suppress campaigns for a random sample and measure the delta. • Tap into platform lift studies. Meta and Google both offer tools to run controlled tests within their ecosystems. If you're working across multiple channels, marketing mix modeling (MMM)—or econometrics—can also help. It's slower and, yes, more complex, but it's useful for spotting directional trends and understanding the broader impact when clean tracking isn't possible. None of these methods are perfect, but they don't need to be. Even directional insight can show you where spend is creating value and where it's not, and that's far more useful than another line on a ROAS report. Leading The Shift Measurement should propel, not paralyze. As a CMO, it's your responsibility to champion the mindset behind incrementality—not as a technical exercise but as a leadership stance. That means setting the expectation that your team, agencies and partners look beyond short-term efficiency and bring forward insight that shows causality, not just correlation. Keep the dashboards—just question what they're really telling you. Here's what that looks like in practice: • Treat ROAS as a signal, not a verdict. It can point to patterns, but it shouldn't be your only lens. • Ask your team for incremental impact. Even rough or directional data is better than chasing false precision. • Champion a test-and-learn culture. Reward smart questions and open-minded analysis—even when the results sting a little. • Protect long-term investments. Not everything will show an immediate return. That doesn't mean it's not working. Sometimes, the things that matter most are the hardest to measure. This isn't about having one perfect metric. It's about building a culture that cares more about truth than theater. Conclusion In a competitive market, the way you measure success defines the way you grow. If your team is chasing ROAS and your competitors are measuring what really works, you're falling behind—whether you can see it yet or not. The shift from attribution to causality isn't a technical upgrade. It's a strategic one. It's about moving from surface-level confidence to real, actionable insight. By embracing incrementality, CMOs can lead smarter teams, make bolder decisions and build more resilient brands in a world where clarity is harder—and more valuable—than ever. Ultimately, if you're measuring ROAS and your competitors are measuring incremental impact, you will plateau, stagnate and eventually lose. Forbes Communications Council is an invitation-only community for executives in successful public relations, media strategy, creative and advertising agencies. Do I qualify?
