Digital Health

AI powered insulin delivery system set for clinical trial at UVA

Managing Type 1 diabetes (T1D) is a constant balancing act, requiring individuals to meticulously monitor their blood sugar levels and administer insulin as needed. The complexity of this task often places a significant mental and physical burden on those living with the condition. To address these challenges, researchers at the University of Virginia (UVA) are embarking on a groundbreaking clinical trial to test an artificial intelligence (AI)-powered device designed to enhance insulin delivery and simplify diabetes management.

A Cutting-Edge Clinical Trial

The trial, spearheaded by leading faculty members from UVA’s School of Data Science and the UVA Center for Diabetes Technology, seeks to evaluate a novel AI-driven feature known as the “Bolus Priming System with Reinforcement Learning” (BPS_RL). This feature, integrated into an advanced Automated Insulin Delivery (AID) system, aims to improve glucose regulation, particularly during meals and overnight, while reducing the need for user intervention.

The research team includes:

• Heman Shakeri, Assistant Professor of Data Science
• Boris Kovatchev, Founding Director of the UVA Center for Diabetes Technology, Professor at the School of Medicine, and Professor of Data Science (by courtesy)
• Anas El Fathi, Research Assistant Professor at the Center for Diabetes Technology and Assistant Professor of Data Science (by courtesy)

Postdoctoral researcher Ali Tavasoli played a crucial role in fine-tuning the BPS_RL system, creating the computer simulation that contributed to the device receiving approval from the U.S. Food and Drug Administration (FDA) for clinical evaluation.

How the Technology Works

The fully automated BPS_RL system functions as an enhancement to AIDANET, an existing diabetes management network comprising:

• A smartphone application that processes glucose data and insulin dosing algorithms
• A Dexcom continuous glucose monitor (CGM) that tracks real-time blood sugar levels
• A Tandem insulin pump that delivers insulin accordingly

By integrating reinforcement learning into this setup, the new technology dynamically adjusts insulin administration in response to real-world data. This allows for adaptive, personalised treatment without requiring continuous user input. The primary aim is to improve blood sugar stability, particularly around mealtimes and overnight, when fluctuations are most difficult to manage.

The Structure of the Clinical Trial

Set to commence in March, the trial will assess the system’s effectiveness over a three-week study period involving 16 adult participants who have previous experience using an AID system. The trial is structured as follows:

• Week 1: Participants will use the standard AIDANET system at home to establish a baseline for comparison.
• Week 2: Participants will stay in a supervised testing location, alternating between the current and AI-enhanced systems in 18-hour monitored sessions.
• Week 3: Participants will return home and use the BPS_RL-enhanced system while being remotely monitored.

To ensure robust results, half of the participants will start with the standard system before switching to the AI-enhanced version, while the other half will follow the reverse order. This crossover design will help researchers compare the two systems’ efficacy in real-world conditions.

Addressing Key Challenges in Diabetes Management

Managing T1D involves numerous variables, including food intake, physical activity, stress, and hormonal fluctuations. Traditional insulin delivery systems require user input, adding complexity and potential for miscalculations. Additionally, AID systems can be expensive and inaccessible to many.

By incorporating AI-driven adaptability, UVA’s new approach seeks to make insulin delivery more effective, less labour-intensive, and potentially more affordable. If successful, this technology could significantly reduce the burden of constant blood sugar management, offering greater peace of mind to those living with diabetes.

A Transformative Step in Diabetes Care

Professor Heman Shakeri emphasised the broader impact of this research, stating:

“This trial isn’t just about advancing technology – it’s a bold step toward transforming diabetes care and uplifting lives. We are committed to creating a fully automated, intelligent insulin delivery system that redefines diabetes management, making treatment simpler, more reliable, and entirely effortless for patients.”

Beyond optimising blood sugar control, the research team envisions a future where AI-powered insulin delivery systems reduce both the mental and financial burdens associated with diabetes care. By making these systems more adaptive, precise, and cost-effective, UVA is paving the way for a more accessible and equitable future in diabetes management.With the trial set to begin soon, the findings could mark a significant leap forward in AI-assisted healthcare, demonstrating how technology can be harnessed to improve lives and reshape the future of chronic disease management.

US bill proposes allowing AI to prescribe medications

A newly proposed bill in the United States Congress could pave the way for artificial intelligence (AI) to legally prescribe medications. If passed, the legislation would mark a significant shift in the integration of AI and machine learning technologies into medical practice, potentially expanding access to prescriptions for individuals who face challenges in securing timely healthcare appointments.

The bill, introduced by Representative David Schweikert (R-Ariz.), aims to amend the Federal Food, Drug, and Cosmetic Act, clarifying that AI systems could qualify as practitioners eligible to prescribe medications. However, these AI-driven prescribing systems would need to be both authorised by the respective state and approved by the U.S. Food and Drug Administration (FDA).

The proposed legislation was introduced in the U.S. House of Representatives in January and subsequently referred to the House Committee on Energy and Commerce for further discussion.

A Premature Move? Experts Express Caution

Despite the bill’s potential to address gaps in care, many medical experts believe that AI is not yet sophisticated enough to take on the role of a prescriber. Dr Adam Rodman, MD, MPH, a hospitalist and director of AI programmes at Beth Israel Deaconess Medical Center, as well as an assistant professor at Harvard Medical School, expressed scepticism about the current feasibility of AI-driven prescribing.

“The technology is not nearly where it needs to be for this kind of prescribing,” Rodman told MedPage Today. However, he acknowledged the rapid pace of advancements in AI. “Things are accelerating so quickly,” he said, adding, “I don’t doubt that we will be having this conversation” in the near future.

Rodman believes the bill reflects a broader enthusiasm for AI technology and its potential to bridge existing healthcare gaps. With many individuals facing difficulties in securing medical appointments, AI could, in theory, provide a solution.

However, significant concerns remain. Dr Stephan Fihn, MD, MPH, an internal medicine physician and professor at the University of Washington, warned that such a system would require an extensive regulatory framework to ensure safety and efficacy.

“There would have to be huge sets of regulations that govern this,” Fihn told MedPage Today. He pointed out that the bill does not currently specify which types of medications AI would be permitted to prescribe, in what settings, or whether human physician oversight would be required.

The Expanding Role of AI in Medicine

AI is already being applied in multiple areas of medicine, with its applications growing at a rapid pace. However, Fihn, who also serves as executive deputy editor of JAMA Network Open, believes that allowing AI to prescribe medications is a premature step.

“The application of AI and machine learning technologies, as described in the bill, seems premature,” he stated. “It appears to enable the actual prescribing of drugs, some of which are very low risk, and some of which are very high risk.”

He compared the issue to the introduction of autonomous vehicles, which have faced high regulatory scrutiny. AI-driven prescribing, he predicted, would likely be subject to similarly stringent standards. “My suspicion is that there will be a very high bar for approving these, at least initially,” he said.

Despite his reservations, Fihn ultimately believes AI prescribing could be a reality in the future. “I think it will come to be,” he said. “And to be honest, properly developed, properly tested, properly managed, and highly regulated, it could be a good thing. But this has to be proven and shown.”

The Legislative Landscape and AI’s Role in Healthcare

Representative Schweikert has yet to provide further comment on the bill, but he has previously expressed strong support for AI’s role in improving healthcare efficiency. In an interview last year with Nextgov/FCW, he argued that AI could be instrumental in reducing government costs.

“Technology has to be part of the way we bend the borrowing and debt curve,” Schweikert stated, highlighting how the broader adoption of AI could “make government better, faster, cheaper” in meeting Americans’ needs.

He also pointed to AI’s growing role in improving diagnostics and optimising both administrative and clinical operations in healthcare. “It’s here, we now just have to build the infrastructure around it,” he said, describing the emergence of a “new medical landscape.” However, he also acknowledged that “technology is starting to move much faster than our regulatory rules.”

This is not Schweikert’s first attempt at pushing for AI-driven prescribing. In January 2023, he introduced a similar bill with the same objective, but it ultimately did not advance out of committee.

A Future of AI-Driven Healthcare?

AI’s role in healthcare continues to evolve, and discussions around its potential are ongoing. In February 2024, a panel hosted by the Kaiser Family Foundation explored AI’s role in simplifying prior authorisation processes. Experts on the panel stressed the importance of “complete transparency” in AI’s use within the healthcare system.

The American College of Physicians has also weighed in on the matter. In a policy position paper published in June 2024, the organisation stated that AI should “complement, not supplant” the decision-making of physicians and other clinicians. This stance underscores the broader concern that AI should serve as an assistive tool rather than replacing human expertise.

As Schweikert’s bill moves through the legislative process, it remains unclear how much support it will garner or whether it will ultimately become law. The concept of AI prescribing medication may have seemed like science fiction only a few years ago, but the landscape is shifting.

“This isn’t sci-fi,” Rodman noted. “It’s just we aren’t there yet.”

Electronic symptom monitoring improves cancer care and reduces emergency visits

A pioneering study has revealed that people with metastatic cancer who actively report their symptoms via a home-based electronic monitoring system experience significant improvements in quality of life, clinical outcomes, and overall well-being. Furthermore, these individuals require fewer emergency department visits compared to those who do not engage in symptom reporting. However, overall survival rates remained comparable between both groups. These findings, led by researchers at the University of North Carolina (UNC) Lineberger Comprehensive Cancer Center, highlight the potential of electronic symptom monitoring to enhance cancer care by facilitating timely interventions.

The results of this extensive multicentre study were published in Nature Medicine on 7 February.

Bridging the Gap in Symptom Management

“Doctors and nurses are often unaware of symptoms and side effects that can worsen for cancer patients between office visits, leading to complications and unnecessary suffering. The patient-reported outcomes, or PRO, system was developed to enable patients to report their own symptoms and side effects and our study showed that PRO keeps care teams informed so that they can intervene promptly to help patients,” stated Dr Ethan Basch, MD, MSc, lead author of the study, Richard M. Goldberg Distinguished Professor of Medicine, and Chief of Medical Oncology at UNC School of Medicine, as well as Director of the Cancer Outcomes Research Programme at UNC Lineberger.

The PRO-TECT randomised clinical trial, conducted across 52 community oncology practices in 26 states, aimed to assess the real-world impact of electronic PRO symptom monitoring on clinical outcomes in comparison to standard care.

A total of 1,191 patients were enrolled in the study. Participants were randomly assigned into two groups: 593 individuals were placed in the PRO arm, where they could report their symptoms through a web-based programme or an automated telephone system, while 598 participants were assigned to the usual care/control arm.

The demographic breakdown of the study participants revealed that the median age was 63 years, with nearly 17% having never used the internet and approximately 26% receiving treatment at rural oncology practices.

Key Findings: Improved Quality of Life and Fewer Emergency Visits

While the primary outcome—overall survival—remained unchanged between the two groups, the use of electronic PRO reporting demonstrated multiple significant benefits:

• Reduction in Emergency Visits
  • Patients in the PRO group experienced a 6.1% decrease in emergency department visits compared to those in the usual care group.
  • Additionally, individuals in the PRO arm experienced a 16% longer duration before their first emergency visit compared to those in the usual care arm.
• Delayed Deterioration in Physical Function and Symptoms
  • The median time before physical function decline was 12.6 months in the PRO group versus 8.5 months in the control group.
  • Participants in the PRO arm exhibited a 31% delay in the worsening of symptoms, with a median progression time of 12.7 months compared to 9.9 months in the control group.
  • Additionally, PRO use was associated with a 28% improvement in health-related quality of life, with median times of 15.6 months versus 12.2 months in the control group.
• Positive Patient Perceptions and Engagement
  • 77% of participants felt PRO enhanced their discussions with their care team.
  • 84% believed PRO helped them feel more in control of their own care.
  • 91% of participants stated they would recommend the PRO system to other patients.
  • The adherence rate was remarkably high, with 91.5% of patients completing scheduled weekly symptom surveys.

Simplifying Symptom Reporting for Patients and Care Teams

One of the major challenges in cancer care has been the burden of electronic medical recordkeeping, which can be frustrating for both physicians and patients. Dr Basch acknowledged this concern, stating:

“Electronic medical recordkeeping has been a bane for some physicians and patients who find it frustrating to navigate patient portal systems—mitigating those frustrations was a central consideration in configuring the PRO system in this study. PRO largely circumvents physicians and is managed by nurses and/or patient navigators, whose jobs often encompass symptom management and care coordination. The PRO systems have also proven very easy for patients to use, in terms of technical ability.”

Generalisability and Future Research

The PRO-TECT trial was intentionally designed to be generalisable across all cancer types and treatment regimens, ensuring that the results apply broadly across diverse oncology populations. The study’s findings confirmed the effectiveness of PRO monitoring across multiple types of cancer, reinforcing its potential for widespread clinical application.

While these results mark the final stage of the PRO-TECT study, researchers have outlined several areas for further investigation. Planned sub-study analyses will explore:

• Patient, nurse, and doctor interviews to gain deeper insights into the experiences of those involved in the trial.
• Outcome variations based on patient characteristics, including cancer type, race, and geographical location.
• Methodological assessments to refine and improve the PRO technology for future implementations.

Despite the strong evidence supporting PRO in metastatic cancer, Dr Basch emphasised the need for further research into its benefits for earlier-stage cancers, concluding:

“While there have been some studies evaluating the benefits of PRO for patients with earlier stage cancers who get chemotherapy, radiation or surgeries for their cancer, there is less conclusive evidence of benefit for PRO in non-metastatic disease, so we certainly hope researchers will take an interest in establishing more studies to provide evidence of that benefit.”

Conclusion

The findings from this study reinforce the importance of proactive symptom monitoring in cancer care. While overall survival remained unchanged, electronic PRO reporting provided notable quality-of-life benefits, delayed symptom progression, and reduced emergency visits. Additionally, patients felt more engaged in their care and valued the improved communication with their healthcare teams. As researchers continue to explore PRO’s broader applications, its potential to transform cancer care and symptom management remains highly promising.

Government launches AI-powered breast cancer screening trial to improve early detection

The UK government has announced the launch of a large-scale trial designed to evaluate the effectiveness of artificial intelligence (AI) in improving early detection rates of breast cancer. The initiative, led by the Department of Health and Social Care (DHSC), will involve nearly 700,000 individuals undergoing breast cancer screening at 30 sites across England. These locations will be equipped with AI-driven technologies aimed at enhancing the accuracy and efficiency of breast cancer diagnoses.

Harnessing AI to Support Radiologists in Early Detection

Breast cancer remains the most prevalent cancer among women in the UK, with approximately 55,000 people diagnosed each year. Early detection is critical for improving outcomes, and the use of AI has the potential to assist radiologists in identifying subtle changes in breast tissue that may indicate cancer.

The technology will analyse mammograms and flag any abnormalities, allowing healthcare professionals to determine whether further investigation is needed. Currently, two specialists are required to review each mammogram; however, AI integration could enable safe and efficient screening with just one specialist, reducing pressure on radiology teams.

A Major Step Forward in Breast Cancer Screening

Professor Lucy Chappell, Chief Scientific Adviser at the DHSC and Chief Executive of the National Institute for Health and Care Research (NIHR), emphasised the importance of the study, stating:

“This landmark trial could lead to a significant step forward in the early detection of breast cancer, offering women faster, more accurate diagnoses when it matters most.”

She also highlighted the role of the NIHR in advancing medical research:

“It is another example of how NIHR research, shaped and funded by the public, is crucial for rigorously testing world-leading new technologies, such as AI, that can potentially save lives while reducing the burden on the NHS.”

The ‘Early Detection using Information Technology in Health’ (EDITH) trial, backed by £11 million in government funding, aims to determine whether AI can reduce the workload of radiologists while maintaining or improving diagnostic accuracy. If successful, this approach could lead to shorter waiting times, improved patient experiences, and earlier interventions for those at risk.

A National Cancer Plan to Drive Improvements in Care

Alongside the AI trial, the government is also seeking input from people with cancer, healthcare professionals, and researchers to shape a new National Cancer Plan. Health Secretary Wes Streeting launched a call for evidence on World Cancer Day, inviting contributions via the Change NHS online platform.

Streeting, himself a cancer survivor, underscored the urgency of improving cancer care in the UK:

“As a cancer survivor, I feel like one of the lucky ones.”

He acknowledged the challenges posed by rising cancer diagnoses and international comparisons:

“With record numbers of people diagnosed with cancer, and Lord Darzi finding that cancer survival is worse in this country than our peers, I know that urgent action is needed to save lives and improve patient care.”

Streeting also reaffirmed the government’s commitment to developing a world-leading cancer strategy:

“That’s why for World Cancer Day, I am committed to publishing a dedicated national cancer plan this year, to unleash Britain’s potential as a world-leader in saving lives from this deadly disease and make the NHS fit for the future through our Plan for Change.”

The feedback gathered from the public and medical professionals will directly inform the development of the 10-year National Cancer Plan, set to be published later this year.

AI and the Future of Healthcare

This initiative aligns with the government’s broader commitment to leveraging AI for healthcare improvements. The AI Opportunities Action Plan, launched in January 2025, has already attracted over £14 billion in investment, reflecting a strong focus on integrating advanced technology into the NHS.

If the EDITH trial proves successful, AI-powered screening could become a transformative tool in breast cancer detection, leading to earlier diagnoses, improved treatment pathways, and ultimately, better survival rates.With AI revolutionising diagnostics and a new cancer plan in development, the UK is taking decisive steps toward enhancing cancer care and ensuring the NHS remains at the forefront of medical innovation.

New guidance issued to strengthen safety and effectiveness of digital mental health technologies

The Medicines and Healthcare products Regulatory Agency (MHRA) has introduced new guidance to ensure that digital mental health technologies are safe, effective, and reliable. As tools such as mobile applications, artificial intelligence-driven assessments, and virtual reality-based therapies become more widely used in both the NHS and private healthcare, the guidance aims to provide clarity for developers on regulatory requirements.

Rob Reid, Deputy Director of Innovative Devices at the MHRA, emphasised the growing potential of digital solutions in mental health support, stating that “effective and acceptably safe digital tools have huge potential to improve mental health support, making help more accessible than ever.” However, he also acknowledged the need for clear regulatory pathways to maintain public trust in these technologies. He explained that the guidance was developed to “support safe access to these important tools by clarifying when a product needs regulatory approval and the steps developers must take.” Ensuring transparency and consistency in regulatory standards, he added, will help to guarantee that “the public can trust these technologies and benefit from the safe, effective mental health support they can provide.”

Clarifying Regulatory Standards for Digital Mental Health Technologies

Published on 3 February 2025, the MHRA guidance sets out how medical device regulations apply to software-based mental health interventions. It provides essential information for manufacturers, detailing which products fall under regulatory oversight, how they should be assessed, and the evidence required to demonstrate their safety and efficacy.

The guidance is one of the key outputs of a three-year Wellcome-funded project launched in 2023, which explored how digital mental health products should be regulated. It was developed in collaboration with the National Institute for Health and Care Excellence (NICE), NHS specialists, researchers, healthcare professionals, and individuals with lived experience of mental health conditions. The goal, according to the MHRA, is to create a framework that balances clinical effectiveness with real-world usability, ensuring that digital mental health technologies are both accessible and safe.

Mark Chapman, Director of HealthTech at NICE, highlighted the importance of this initiative, noting that “providing more detailed guidance to the developers of digital mental health technologies helps us to ensure that technologies being considered for NICE assessments have received an appropriate level of regulatory scrutiny to assure their safety.” He stressed that, with a wide range of digital mental health products on the market, “it is important people can understand how regulations apply to different products.” The new guidance, he explained, will also help NICE refine its own evaluations: “This guidance will help inform our evaluations and ensure that NICE is able to publish useful, usable, and timely guidance that allows people with mental health conditions to access safe and effective innovations faster.”

Supporting Compliance for Market-Ready Digital Tools

The MHRA is encouraging all manufacturers of digital mental health technologies to carefully review the guidance to ensure compliance before launching their products. Regulatory approval, according to the agency, is essential not only for patient safety but also for maintaining public confidence in the effectiveness of digital mental health tools.

Professor Miranda Wolpert, Director of Mental Health at Wellcome, underlined the significance of digital solutions in addressing global mental health challenges. She pointed out that “with millions of people around the world held back by mental health problems, digital mental health therapies have huge potential to be scalable and accessible.” However, she acknowledged the difficulties in finding the right balance between ensuring safety and avoiding excessive regulation. “It is not easy to navigate between over and under regulation in this area,” she noted. Given the rapid evolution of digital mental health technologies, she described the new guidance as a practical step forward, stating that “in a fast-moving and continuously evolving digital space, these thoughtful guidelines appear well positioned to strike a pragmatic balance between making digital mental health technologies accessible to those with a range of mental health needs whilst also ensuring they are safe, effective and as transparent as possible.”

Upcoming Changes to Post-Market Surveillance Regulations

Beyond the new guidance for digital mental health technologies, the MHRA has also issued fresh regulatory directives to help manufacturers prepare for upcoming changes in post-market surveillance (PMS) regulations. These changes, set to take effect on 16 June 2025 across England, Scotland, and Wales, are intended to enhance the monitoring of medical devices after they reach the market, ensuring ongoing compliance with safety and performance standards.

As digital mental health technologies continue to play an increasingly central role in mental healthcare, the MHRA’s latest guidance represents an important step towards ensuring that individuals have access to safe, high-quality digital interventions. By clarifying regulatory expectations and reinforcing public confidence, the guidance aims to support the development of innovative, effective, and secure mental health technologies that can meet the needs of people seeking support.

Government launches ‘Humphrey’ AI tool to transform NHS and public services

The UK government has unveiled a new suite of artificial intelligence (AI) tools, collectively nicknamed ‘Humphrey’, designed to streamline operations within the NHS and other public services. The initiative is part of a broader effort to modernise the country’s digital infrastructure and deliver more efficient, cost-effective public services.

Developed by a team of expert AI engineers at the Department for Science, Innovation and Technology (DSIT), the ‘Humphrey’ bundle draws its name from the iconic fictional civil servant Sir Humphrey Appleby, made famous in the BBC political satire Yes, Minister. The tools aim to tackle inefficiencies, reduce bureaucracy, and enhance productivity across government departments and public sector organisations.

The package comprises several specialised tools, including:

• Consult: Designed to analyse responses from public consultations, making the process of gathering and evaluating feedback more efficient.
• Minute: An AI-powered meeting transcription service that generates accurate and detailed summaries of discussions.
• Redbox: A generative AI tool to assist civil servants with everyday tasks, such as summarising complex policies, preparing briefings, and drafting documents.
• Lex: A research tool that enables officials to navigate and interpret legal texts more effectively.

Empowering People Through Digital Transformation

Wes Streeting, the Health Secretary, emphasised how these technological advancements are pivotal to rebuilding the NHS and empowering individuals to take control of their own healthcare.

“We are bringing our analogue NHS into the digital age. Our Plan for Change will rebuild our NHS, put patients in control of their own healthcare, and arm staff with the latest groundbreaking technology, ending the needless bureaucracy faced by patients up and down the country,” Streeting said.

He highlighted the government’s ongoing efforts to improve digital healthcare delivery, including updates to the NHS App. “We’ve already set out plans to transform the NHS App so patients can choose providers and book appointments, and we’re harnessing artificial intelligence to deliver faster and smarter care across the country,” he added.

By embracing these innovations, the government hopes to achieve dual objectives: improving the efficiency of public services while delivering substantial savings for taxpayers. “By embracing technological advancements, we can both make substantial savings for the taxpayer and build a health service fit for the 21st Century,” Streeting concluded.

Enhancing Digital Services Across Government

In addition to modernising healthcare, the government plans to overhaul the way it manages digital services across all public sectors. Central to this effort is the creation of a Digital Commercial Centre of Excellence, which will focus on improving the government’s £23 billion annual spend on technology.

This new centre will promote collaboration across public sector organisations, enabling them to negotiate contracts collectively, thereby reducing costs. It will also open up new opportunities for UK-based start-ups and scale-ups to contribute to economic growth and job creation. These efforts align with the Prime Minister’s Plan for Change, which prioritises economic growth and innovation.

The government will publish a digital and AI roadmap later this summer, outlining its strategic approach to leveraging technology across public services. This roadmap will coincide with the second phase of the ongoing spending review and will detail specific priorities for AI and digital transformation.

A Common-Sense Approach to Data Sharing

To tackle inefficiencies in data sharing, the government is committed to adopting a more pragmatic and transparent approach. A recent press release explained that central government departments will share data more seamlessly with local councils to tackle fraud and support businesses.

The government acknowledged that it inherited “a dire system which over relies on ways of communicating that should be left in the last century.” Improved data sharing and collaboration are expected to reduce operational delays and improve service delivery.

Peter Kyle, the Science Secretary, underscored the urgency of this digital overhaul. “Sluggish technology has hampered our public services for too long, and it’s costing us all a fortune in time and money,” Kyle stated.

He also highlighted the frustrations that outdated systems impose on individuals. “Not to mention the headaches and stresses we’re left with after being put on hold or forced to take a trip to fill out a form,” he said.

Kyle reiterated the government’s commitment to using AI to deliver its Plan for Change. “My department will put AI to work, speeding up our ability to deliver our Plan for Change, improve lives and drive growth. We will use technology to bear down hard on the nonsensical approach the public sector takes to sharing information and working together to help the people it serves,” he added.

Delivering on Expert Recommendations

The introduction of ‘Humphrey’ delivers on the recommendations of the AI Opportunity Action Plan, which calls for the rapid development, testing, and implementation of tools that enhance public sector productivity and improve services for the population.

As the government continues to embrace technological innovations, these efforts represent a significant step toward creating a public service infrastructure that meets the demands of the 21st century, with an emphasis on empowering people and improving their experience of government services.

Remote patient monitoring and mobile health powering the digital health revolution

Digital healthcare is undergoing a profound transformation, largely propelled by the adoption of remote patient monitoring and mobile health (mHealth), according to a new report. The 2024 edition of GlobalData’s Digital Health report highlights a seismic shift from traditional pen-and-paper practices to a fully digitised healthcare system—a transition accelerated by the Covid-19 pandemic.

“Digital methods of care were introduced rapidly to accommodate for the number of patients and the need to social distance,” the report notes. “Since then, digital health methods have advanced and continue to, changing how patients receive care.”

The report underscores that digital health innovations not only improve healthcare delivery but also have the potential to reduce inefficiencies, increase accessibility, lower costs, enhance quality, and personalise care. It predicts that technologies such as artificial intelligence (AI), machine learning, and big data will continue to drive advancements in digital health, particularly in areas like remote patient monitoring and mobile health.

Remote Patient Monitoring: Enhancing Care While Reducing Costs

One of the most transformational trends, remote patient monitoring, enables the continuous collection and transmission of health data through wireless, often wearable devices. This approach allows individuals to remain at home while healthcare professionals receive real-time updates on their health.

“Remote patient monitoring devices were one of the fastest-growing and most in-demand industries in 2020 and 2021,” the report states. “They have demonstrated their potential to improve patient care, reduce readmissions, and facilitate early discharge.”

This technology not only enhances efficiency by minimising the need for in-person follow-ups but also empowers individuals to better understand and manage their health conditions. For example, patients with chronic diseases can benefit from early intervention based on data trends, reducing the likelihood of hospitalisation and enabling tailored care plans.

Mobile Health (mHealth): From Wellness to Professional Healthcare Applications

The role of mHealth has similarly expanded, encompassing everything from consumer wellness apps to tools used by healthcare professionals. The Covid-19 pandemic further accelerated the adoption of mHealth applications, which now include functionalities like health tracking, appointment scheduling, disease monitoring, and treatment facilitation.

“Similar to remote patient monitoring devices, mHealth apps and electronic medical record systems have been adopted quite quickly, in part due to the Covid-19 pandemic,” the report explains. “While the use of mHealth apps has increased quickly in recent years, they have been around for quite a while, especially health and wellness mHealth apps (calorie counters, fitness trackers, sleep trackers, etc.).”

Today, mHealth apps cater to both consumers and clinicians, with increasing adoption of apps that support disease management, symptom tracking, and even diagnostic testing. This dual utility reflects the growing role of mHealth in bridging gaps between patients and healthcare systems, enhancing both accessibility and engagement.

The Future of Digital Healthcare

Looking ahead, the report anticipates that advancements in AI and machine learning will further refine the capabilities of digital health technologies. These innovations are expected to enable predictive analytics, optimise treatment plans, and improve user experiences for both patients and healthcare professionals.

“Digital health will continue to transform healthcare by leveraging technologies like remote patient monitoring, mobile health, big data and more,” the report concludes. “Digital health will also continue to evolve through artificial intelligence and machine learning advancements that will enhance the various components of digital health devices.”

Digital heart twins have potential to deliver safer faster heart rhythm treatments

Advances in medical technology are paving the way for safer and more effective treatments for individuals living with life-threatening heart rhythm disorders. A recent study published in the American Heart Association journal Circulation highlights the groundbreaking potential of using computer-generated digital heart twins to identify problematic areas deep within the heart muscle. This innovative approach offers a non-invasive alternative for addressing scar-related heart rhythm abnormalities, often linked to previous heart attacks or genetic conditions.

The research team demonstrated that digital heart twins could effectively pinpoint scarred regions in the heart responsible for abnormal rhythms, potentially enabling cardiologists to deliver faster, more targeted treatments. “People are now living with the consequences of heart attacks for many years, so the number of people who need procedures to treat these life-threatening abnormal heart rhythms is rising,” explained Dr Michael Waight, the lead author of the study and a cardiology registrar at St George’s University of London. “If digital twins were to become a reality, it could offer a safer and potentially more effective means of treatment.”

Scar-dependent ventricular tachycardia (VT) is a severe and sporadic heart rhythm condition caused by scar tissue on the heart muscle. It is treated either by implanting a defibrillator to restore normal rhythm during episodes or by performing catheter ablation, a procedure that involves burning scar tissue to prevent further abnormal rhythms. However, these treatments have notable limitations. Defibrillators, while life-saving, do not prevent rhythm recurrences and may cause discomfort from repeated shocks. Ablation procedures, on the other hand, face challenges in locating and treating deeply embedded scarred areas, particularly when scarring occurs at multiple sites.

To identify scarred regions, clinicians typically map the heart’s interior by inserting a catheter to detect abnormal electrical signals. In some cases, they induce irregular rhythms to locate the source of the problem before ablating the affected areas. This process can be time-intensive, carries risks, and may lead to surgical complications or incomplete treatment if all problem sites are not identified.

“It is quite a time-consuming process that isn’t without risk,” said Dr Waight. “These patients can be quite unwell, with poor heart function, and this is a long and arduous procedure. We’re looking at ways we can improve that by hopefully trying to shorten the procedure, make it more accurate and more targeted to where the problem is.”

The study explored whether digital heart twins could address these challenges. By using enhanced cardiac imaging and other clinical data, researchers created computer models replicating the structure and function of the hearts of 18 individuals undergoing catheter ablation for scar-related VT. These models were then tested for electrical rhythm abnormalities similar to those observed in the patients.

The findings were promising. Areas flagged by the digital twins as problematic showed a 41% higher frequency of electrical abnormalities compared to unflagged regions. Moreover, the models successfully predicted approximately 80% of sites with slowed electrical signals, typically found near scarred heart tissue. “Digital twins can render the heart in 3D and see exactly where the faulty circuit is,” explained Dr Waight. “That means we can have an idea of the area to target before the patient even comes to the catheter lab facility. We already know where we need to go and don’t need to spend hours making a map of the heart.”

This predictive capability could significantly reduce procedure times and recurrence rates, minimising the need for additional treatments. “The nature of the digital twin is that it doesn’t just predict VTs the patient is having now, but also all possible VTs that might occur in the future,” Dr Waight added.

However, the researchers emphasised that the technology’s clinical application is still in its early stages. “The first step was to prove these places in the heart are important, compared to other places not picked up by the digital heart. The next step would be a clinical trial where we compare the current standard of care of VT ablation to a strategy where we are guided by the digital twin from the outset,” said Dr Waight.

Experts unaffiliated with the study expressed optimism about the technology’s potential. Dr Dhanunjaya Lakkireddy, executive medical director of the Kansas City Heart Rhythm Institute, noted that using digital twins could improve patient outcomes by enhancing procedural precision. “If you are able to successfully eliminate these areas that are potentially sites of ventricular tachycardias, you can reduce the time of the procedure and go precisely to the areas that are important with minimal unnecessary ablation,” he said. “This could potentially improve overall morbidity and mortality with a higher success rate and result in dramatically improved patient outcomes.”

Dr Lakkireddy also highlighted the innovative nature of the approach: “The digital heart twin gives you a road map of areas to focus on. It’s an incredibly powerful advancement. It’s really moving the field forward, making these procedures a lot more effective and more precise.”

Despite its promise, the technology faces challenges related to cost and accessibility. “Whether we can apply this on a large scale is an open question,” said Dr Lakkireddy. “Creating a digital twin is a very expensive process at this time.”

This pioneering research lays the foundation for transforming the treatment of scar-dependent VT. By combining advanced imaging with predictive modelling, digital heart twins could herald a new era of safer, more effective cardiac care, improving the lives of countless individuals managing serious heart rhythm disorders.

Transforming obesity care: New Warwick study explores digital based weight management support

Researchers at the University of Warwick are conducting a pioneering study to explore how digital tools, including a mobile app and remote consultations, could help the NHS provide better support for individuals living with obesity. This ambitious project has the potential to transform how care is delivered and make weight management services more accessible and equitable.

The study, funded by the National Institute for Health and Care Research (NIHR), will compare the effectiveness of a digital care pathway incorporating the Gro Health W8Buddy app with traditional NHS specialist weight management services. By improving access to care and addressing health inequalities, the research aims to benefit people from all walks of life.

Tackling a National Health Crisis

Obesity affects over a quarter of the UK population and remains one of the country’s most pressing health issues. The condition is linked to serious complications, such as type 2 diabetes, cardiovascular disease, and reduced quality of life. However, access to NHS weight management services is inconsistent, leaving many people waiting for support.

Dr Petra Hanson, Clinical Lecturer at Warwick Medical School and lead researcher for the study, emphasised the urgent need for innovation, “This is about finding innovative ways to improve access and ultimately health outcomes for different groups of people, including those from diverse age groups and ethnic backgrounds. With thousands of people on waiting lists for specialist weight management services, we need to change the way we work.”

Study Details and Objectives

The study will follow 450 participants across four hospital sites, collecting data on outcomes such as weight loss, quality of life, speed of treatment, healthcare resource usage, and overall health improvements. Researchers will track these outcomes over 18- and 24-month periods to determine the effectiveness of the digital and traditional care pathways.

Dr Hanson highlighted the role of digital tools in modern healthcare, “We want to know if digital pathways like W8Buddy can be incorporated into NHS treatment pathways. We don’t want to replace traditional care but instead use digital tools to improve choice and augment the way we work. This isn’t just about the technology; it’s about giving people choice and evidence-based tools.”

A Patient-Centred Approach

The study places individuals living with obesity at the heart of its design. Participants will contribute by sharing their experiences through surveys, interviews, and focus groups. This feedback will guide researchers in tailoring the digital care pathway to meet individual needs, ensuring the system is user-friendly, effective, and widely accessible.

By prioritising patient and public involvement, the researchers aim to create a service that resonates with users and offers a practical solution to a complex health challenge.

Collaboration and Diversity

The project is a collaboration between the University of Warwick, University Hospital Coventry and Warwickshire, and digital health company DDM Health. Recruitment sites include hospitals in Coventry, London, Birmingham, and Wales, selected for their diverse patient populations. This ensures the research captures a broad spectrum of experiences, enhancing its relevance and applicability.

Implications for NHS Policy

The findings of this study could have far-reaching implications for NHS policy. By demonstrating how digital solutions like W8Buddy can be integrated into NHS care, the research has the potential to reshape weight management services across the UK.

If successful, the digital care pathway could provide a scalable, cost-effective option for supporting people living with obesity. It would offer individuals greater flexibility in managing their health while complementing existing services.

Looking Ahead

This innovative study could signal a new era in obesity care, bridging gaps in access and equity while embracing the potential of digital health tools. As Dr Hanson remarked, “This isn’t just about technology; it’s about improving health outcomes and giving people the tools they need to succeed.”

By combining cutting-edge technology with a patient-first approach, this research promises to offer hope and empowerment to thousands of people across the UK, potentially transforming the future of obesity care in the NHS.

AI innovation offers hope for safer maternity services

Researchers at Loughborough University have developed an innovative artificial intelligence (AI) tool designed to enhance safety in maternity care by identifying the human factors that could impact outcomes. Known as I-SIRch (Identifying Social, Technical, and Organisational Factors in Incident Reports), this tool was created by Professor Georgina Cosma and Professor Patrick Waterson to provide new insights into areas such as communication, teamwork, and decision-making within maternity services.

The findings from the analysis of 188 real maternity incident reports, which the tool was trained on, were published in the International Journal of Population Data Science on 20 November 2024. The researchers hope the tool will help healthcare providers design targeted interventions to improve outcomes for mothers and babies.

Highlighting the importance of this work, Professor Waterson explained, “Our work opens up new possibilities for understanding the complex interplay between social, technical, and organisational factors influencing maternal safety and population health outcomes.” He noted that the research addresses the gaps identified in the 2022 Ockenden Review, which found that “repeated failures” in NHS maternity care led to over 200 preventable infant deaths between 2000 and 2019. “By taking a more comprehensive view of maternal healthcare delivery, we can develop targeted interventions to improve maternal outcomes for all mothers and babies,” he added.

Addressing Challenges in Incident Reviews

Traditionally, when adverse incidents occur in maternity care, investigations require manual reviews of reports, which are time-consuming and prone to inconsistencies due to reliance on individual interpretation. I-SIRch automates and standardises this process, allowing for the analysis of multiple reports simultaneously to identify recurring human factors.

In testing, I-SIRch successfully identified human factors in each report and provided precise insights into areas where additional support could improve outcomes. This capability could revolutionise how incident investigations are conducted, helping allocate resources more effectively and preventing future adverse events.

Discussing the broader implications of this work, Professor Cosma stated, “We are seeking to collaborate with hospitals, healthcare organisations, and investigation bodies to further refine and apply our AI tool to reports. These partnerships will help us extract vital intelligence to prevent adverse incidents and ensure the safety of all mothers and babies.” She also emphasised the potential for I-SIRch to be adapted for other applications, explaining, “We also hope to adapt the tool for use with other types of reports, such as adverse police incident reports, where understanding the human factors involved can help prevent future incidents and improve response strategies.”

Future Development and Tackling Inequities

The I-SIRch project, jointly funded by the Health Foundation and the NHS AI Lab at the NHS Transformation Directorate, with support from the National Institute for Health and Care Research, has achieved significant milestones. However, the researchers are now seeking additional funding to refine the tool further. This includes incorporating a larger and more diverse dataset, which will allow I-SIRch to address the unique challenges faced by mothers from ethnic minority groups, who often experience disparities in maternity care.

Dr Jonathan Back, a safety insights analyst at the Health Services Safety Investigations Body, highlighted the potential of I-SIRch, saying, “The AI tool could help analysts working in health and care to identify where there are inequalities, maximising learning by bringing together findings from multiple investigations.”

The Role of AI in Transforming Maternity Care

The potential of AI in improving maternity care has been further underscored by a 2024 University of Birmingham review, which examined over 12,000 papers and 87 articles. The study found that AI software improved the odds of women receiving good maternity care by an impressive 69%.

Through tools like I-SIRch, healthcare providers can gain valuable insights into human factors, enabling them to implement more precise and equitable interventions. By addressing the systemic challenges in maternity services, this innovative approach holds the promise of safer, more effective care for all families.

Oracle unveils groundbreaking AI-powered electronic health record system

Oracle has announced the launch of a cutting-edge electronic health record (EHR) system, marking its most significant healthcare innovation since acquiring Cerner for $28 billion in 2022. This latest development combines artificial intelligence (AI) with cloud technology, aiming to transform the way clinicians interact with patient data and enhance overall healthcare delivery.

A New Era for Electronic Health Records

EHRs are digital repositories of individuals’ medical histories, maintained and updated by healthcare professionals over time. While vital to modern healthcare systems, many EHR platforms have been criticised for their complexity and inefficiency, often taking valuable time away from patient care. Oracle’s new EHR seeks to address these challenges by introducing a more intuitive, AI-powered solution.

The new system eliminates traditional menus and drop-down screens, allowing clinicians to retrieve patient information by using voice commands. This innovation is designed to simplify workflows and enable healthcare professionals to dedicate more time to patient interactions.

“It’s not just a scribe. It’s not an assistant. It’s almost like having your own resident,” explained Seema Verma, Oracle’s Executive Vice President and General Manager of Health and Life Sciences, in an interview with CNBC.

Competing in a Tough Market

Oracle’s push into the highly competitive EHR market comes at a critical time. The company has faced challenges, including significant losses in market share, while Epic Systems, its chief competitor, has strengthened its position. According to KLAS Research, Oracle experienced its largest net hospital loss in 2023, whereas Epic achieved a net gain in acute care market share. Financially, Cerner contributed $5.9 billion to Oracle’s revenue in fiscal 2023, while Epic generated $4.9 billion during the same period.

The development of Oracle’s new EHR began after the acquisition of Cerner, but the system was built independently of Cerner’s legacy infrastructure. As Verma noted, “Just think about crumbling infrastructure in a house; you’re not going to put new things on top of it. That was the conclusion that we came to when we looked at the Cerner technology, so what we’re introducing to the market is something that’s brand new.”

A Glimpse into Oracle’s AI-Driven EHR

Suhas Uliyar, Oracle’s Senior Vice President for Product Management in Clinical and Healthcare AI, provided a virtual demonstration of the new EHR. He highlighted how it streamlines administrative and clinical tasks for healthcare professionals, showcasing its simplicity and functionality.

The browser-based interface opens to a chronological list of appointments and a search bar. Using the microphone icon, clinicians can ask questions like, “How many openings do I have today?” or “What new patients are on my schedule?” The AI provides immediate, precise answers, reducing the need for manual searches.

When a clinician accesses a patient’s chart, they can view AI-generated summaries of the individual’s medical history, recent changes, current medications, lab results, and other essential details. The system also enables specific, voice-activated queries, such as “Has she reported shortness of breath?” or “What antibiotics have been used to treat her urinary tract infections?”

“It’s going through the entire history, all the records, and it gives me a very specific answer,” Uliyar said. “I didn’t have to scroll through 15 different documents to find that.”

The system’s AI learns from clinicians’ habits, adapting to their preferred workflows, commonly prescribed medications, and frequent queries. Even when questions are phrased imperfectly, the AI delivers accurate responses. To ensure trust and transparency, clinicians can access citations and review the original records behind AI-generated answers. Recommendations for medication dosages and other clinical decisions are linked to validated databases.

Integration and the Clinical AI Agent

Oracle’s new EHR incorporates features like the Oracle Health Clinical AI Agent, previously rolled out to existing Cerner customers. This tool automates documentation, allowing clinicians to record patient interactions via an app. The AI then generates clinical notes, reducing the administrative burden. Already adopted by around 70 organisations, the Clinical AI Agent is also being adapted for nursing staff.

This agent is embedded within the new EHR but remains available as a standalone, EHR-agnostic product, ensuring flexibility for diverse healthcare settings.

Customisation and Market Disruption

Oracle plans to launch an early adopter programme for the new EHR next year, working closely with healthcare providers to tailor the system to their specific needs. By transitioning customers to the cloud, Oracle aims to streamline the implementation process.

“We see it as very disruptive to the market,” Verma stated. “Our EHR is going to solve a lot of long-standing problems that we’ve had in health care.”

Oracle’s ambitious vision for its AI-powered EHR has the potential to redefine how clinicians interact with technology, paving the way for more efficient, patient-focused care.

Advanced AI model uncovers high-risk heart failure phenotype in people with diabetes

An artificial intelligence (AI) model has demonstrated its potential to provide a “comprehensive characterisation” of diabetic cardiomyopathy, according to researchers from UT Southwestern Medical Center in Dallas. This breakthrough could significantly advance efforts to prevent heart failure in individuals with diabetes.

The study revealed that a machine learning model can effectively identify individuals living with diabetic cardiomyopathy—a disorder of the heart muscle in people with diabetes that can progress to heart failure.

According to the findings, the AI model is capable of detecting a high-risk phenotype associated with diabetic cardiomyopathy. This insight offers the possibility of earlier interventions, potentially averting the onset of heart failure.

Dr Ambarish Pandey, the study’s first author, explained:
“This research is noteworthy because it uses machine learning to provide a comprehensive characterisation of diabetic cardiomyopathy – a condition that has lacked a consensus definition – and identifies a high-risk phenotype that could guide more targeted heart failure prevention strategies in people with diabetes.

“Phenotypes are observable physical properties of individuals that give them specific biological traits.”

Study Design and Key Findings

The study involved an analysis of health data from 1,000 adults participating in the Atherosclerosis Risk in Communities cohort. All participants had diabetes but no prior history of cardiovascular disease.

Using this dataset, researchers assessed 25 echocardiographic parameters and cardiac biomarkers, ultimately identifying three distinct patient subgroups.

One of these subgroups, comprising approximately 27% of participants, was classified as the high-risk phenotype group. Members of this group exhibited elevated levels of NT-proBNP, a biomarker indicative of abnormal heart remodelling and cardiac stress.

Importantly, individuals within the high-risk phenotype group had a 12.1% higher likelihood of developing heart failure compared to those in the other two subgroups.

The findings suggest that between 16% and 29% of people with diabetes may exhibit this high-risk phenotype.

AI in Action: A Neural Network for Early Detection

To extend the impact of this research, the team developed a deep neural network classifier capable of identifying more cases of diabetic cardiomyopathy.

Dr Pandey elaborated:
“Clinically, this model could help target intensive preventive therapies, such as SGLT2 inhibitors, to patients most likely to benefit. “It may also help enrich clinical trials of heart failure prevention strategies in people with diabetes.”

Building on Previous Research

Dr Pandey highlighted that this study builds on earlier work examining the prevalence and prognostic implications of diabetic cardiomyopathy in adults living in the community.
“It extends those efforts by using machine learning to identify a more specific high-risk cardiomyopathy phenotype,” he said.

This innovative approach could help redefine how diabetic cardiomyopathy is diagnosed and managed, paving the way for more personalised and effective treatment strategies.

The full study can be accessed in the European Journal of Heart Failure.