Digital Health Coffee Time Briefing ☕

Your morning summary of digital health news, information and events to know about if you want to be “in the know”.

👇 News

👩‍⚕️Researchers from The University of Hertfordshire Integrated Care System partnership are developing an AI model to improve patient outcomes across the East of England. The project uses AI to analyse and forecast healthcare demand, enabling NHS and council managers to make better-informed decisions about resources, staffing, and patient care, that provide cost efficiencies.

☢ The Royal Marsden NHS Foundation Trust has successfully delivered online adaptive radiotherapy (oART) on an Elekta linear accelerator, potentially improving treatment precision and helping to minimise damage to healthy tissue. oART treatments usually require specialised machines, the trust says, but using the Elekta tool shows how they can be delivered on machines already in use across the NHS, opening up the opportunity for better access for patients.

🍺 UK HealthTech startup Nul, a supported alcohol reduction platform that offers access to clinical care and prescription medication, has raised nearly £873,4000 ($1m) in seed funding to support UK launch, team expansion, and international growth plans. The round was led by dmg ventures and BYVP, along with a group of angel investors in the technology, healthcare, and consumer technology sector, EU-Startups reports.

📲 Lancashire and South Cumbria NHS Foundation Trust is introducing EBO’s Intelligent Patient Portal as part of its digital transformation programme to make it easier for people to access services and support mental health and community care. The portal is powered by conversational AI and informed by behavioural science, to provide patients a simple and accessible way to manage their care.

🧬A collaboration between the University of Liverpool and Boston-based biotech company BPGbio aims to accelerate drug discovery using AI following a trade mission led by Liverpool Mayor Steve Rotheram. The partnership, facilitated through the University’s Civic HealthTech Innovation Zone programme, will use advanced AI techniques to analyse large-scale healthcare data, including medical scans, lab results, and biological information about genes and proteins.

🤰Maternal health AI company Matresa has raised £315,000 in pre-seed funding to accelerate the development and rollout of its tailored preventative maternal health platform. The Matresa platform provides support for women throughout the maternal journey, from preconception through pregnancy and into postpartum recovery, and is expected to launch in summer.

❓ Did you know that?

Hundreds of thousands more women attended NHS breast screening last year and thousands more cancers were diagnosed early in England, according to NHS England.

The latest figures show that in 2024/25, 1.94 million women aged 50 to 70 attended screening within six months of invitation – up nearly 200,000 (193,745) from 1.75 million the previous year.

As a result, 19,291 cancers were detected – nine cases in every 1,000 women screened – which is up almost 16% on the previous year, when 16,677 cancers were diagnosed through NHS breast screening.

Attendance among women invited for screening for the first time reached 63.6%, with 4.79 million eligible women now up to date with their breast screening – the highest levels in a decade.

Despite this progress, the latest statistics from the NHS Breast Screening Programme show that around three in ten women did not take up the offer of screening.

Dr Harrison Carter, director of screening at NHS England, said: “Breast screening can save lives. With nearly 20,000 cancers detected early through screening last year, it’s encouraging to see more women attending, especially those invited for the first time, because making screening a habit can help protect your health for years to come.”

Local screening services are working with NHS England to improve uptake, including targeted outreach in lower-attending areas, reminder texts, and expanded use of mobile screening units to bring services closer to home.

📖 What we’re reading

A study discovered that large language models (LLMs) are no better at helping users to identify medical conditions and taking steps to treat them than Google.

The research, published in Nature Medicine on 9 February 2026, tested whether LLMs, a form of AI, can assist members of the public in identifying underlying conditions and choosing a course of action in 10 medical scenarios in a controlled study with 1,298 participants.

Participants were randomly assigned to receive assistance from an LLM (GPT-4o, Llama 3, Command R+) or a source of their choice, mainly internet search engines, as a control group.

Tested alone, LLMs complete the scenarios accurately, correctly identifying conditions in 94.9% of cases and disposition in 56.3% on average.

However, participants using the same LLMs identified relevant conditions in fewer than 34.5% of cases and disposition in fewer than 44.2% – both no better than the control group.

According to the authors, previous work has shown that using LLMs does not improve clinical reasoning in physicians, and the study suggests that this also extends to the public.

They also recommend systematic human user testing to evaluate the interactive capabilities of LLMs before public deployments in healthcare.

🚨 Upcoming events

• 24 and 25 March, Birmingham NEC – Digital Health Rewired 2026