🔍 DataBlast UK Intelligence

London Ambulance Service Pioneers UK's First AI Trial for Emergency Response

Executive Summary: Transformative AI Implementation in Emergency Services

London Ambulance Service has begun trialling Artificial Intelligence (AI) to support paramedics in treating more patients, marking a watershed moment for UK emergency services. The trial, announced on September 5, 2025, uses AI developed by TORTUS, initially tested at Great Ormond Street Hospital before expanding into emergency care settings.

[cite author="Emergency Services Times" source="September 5, 2025"]London Ambulance Service has begun trialling Artificial Intelligence (AI) to support paramedics in treating more patients, marking a first for UK emergency services[/cite]

The significance extends beyond London - this represents the first comprehensive AI deployment in UK ambulance services with measurable productivity gains that could transform national emergency response capabilities.

The TORTUS AI Technology: From Pediatrics to Paramedics

The AI system's journey from Great Ormond Street Hospital to ambulances demonstrates thoughtful healthcare technology migration:

[cite author="GOSH DRIVE Innovation Unit" source="September 2025"]The study, led by GOSH's Innovation Unit, GOSH DRIVE, was conducted across nine NHS sites in London to assess the impact of an AI-scribing tool, TORTUS, which automatically transcribes consultations and drafts summarised clinical notes for clinicians to review[/cite]

The technology serves multiple functions in ambulance settings. Around 20% of 999 callers in London receive treatment over the phone, and most can now benefit from the AI system's documentation capabilities. The technology has also been tested in ambulances, where crews found it saved significant time during face-to-face assessments.

[cite author="GOSH Study Results" source="September 2025"]Between June 2024 and February 2025, more than 17,000 patient encounters were evaluated across sites including hospitals, GP practices, mental health services and ambulance teams[/cite]

Measurable Impact: Quantifying the Transformation

The trial results demonstrate remarkable efficiency improvements across emergency services:

[cite author="TORTUS Trial Analysis" source="September 2025"]Results showed a 23.5% increase in direct patient interaction time during appointments, alongside an 8.2% reduction in overall appointment length when AI-scribes were used[/cite]

For emergency departments specifically, the impact is even more pronounced:

[cite author="Emergency Department Analysis" source="September 2025"]A&E saw particularly strong results, with a 13.4% increase in patients seen per shift[/cite]

These metrics translate to thousands more patients receiving care without additional resources - a critical capability given current NHS pressures.

National Implications: £834 Million Annual Savings Potential

[cite author="Clare McMillan, Chief Digital Officer at London Ambulance Service" source="September 5, 2025"]If every ambulance service adopted this technology, the improvements we are seeing in London would translate into thousands more patients getting faster and better care[/cite]

The financial implications are staggering:

[cite author="NHS Economic Analysis" source="September 2025"]The study found that AI-scribing technology can reduce clinician workload while improving patient care, with potential to unlock £834 million a year if rolled out nationally[/cite]

This represents not just cost savings but fundamental service transformation - enabling existing staff to deliver more care with better outcomes.

Government Response and Strategic Integration

[cite author="Stephen Kinnock, Health Minister" source="September 2025"]This is exactly the kind of innovation we need as we work to build an NHS fit for the future and end hospital backlogs. By freeing up clinicians from administrative burden to spend more time with patients, we're not just improving efficiency – we're enhancing the human connection that sits at the heart of great healthcare[/cite]

The trial's success has already influenced national healthcare strategy:

[cite author="Government Health Policy" source="September 2025"]The trial's success has already helped inform the government's 10-year Health Plan for England, with a strong focus on productivity and patient outcomes[/cite]

The government views this as part of broader NHS digitalization:

[cite author="Health Ministry Statement" source="September 2025"]As part of our 10 Year Health Plan, technologies like AI scribes are crucial in our shift from analogue to digital healthcare[/cite]

Implementation Timeline and Rollout Strategy

The phased approach demonstrates careful planning:

[cite author="NHS England Planning" source="September 2025"]Following the success of the trial, a rollout of AI-scribe technology across outpatient settings at GOSH is planned to begin this autumn[/cite]

The findings have already informed national guidance:

[cite author="NHS England" source="September 2025"]The findings have already informed NHS England's national guidance on AI-enabled scribing and contributed to the Government's 10-year Health Plan for health innovation and productivity[/cite]

Broader Context: UK Ambulance Response Crisis

This innovation arrives at a critical juncture for UK ambulance services. Current performance metrics highlight the urgency:

- Category 2 ambulance response times average 27 minutes 34 seconds (April 2025) against an 18-minute target
- Some regions like South Western Ambulance Service average 51 minutes 45 seconds - nearly triple the target
- Over 360,000 patients attend A&E more than five times yearly, straining emergency resources

The AI implementation directly addresses these challenges by increasing capacity without additional staffing.

Competitive Landscape: UK Leading Global Innovation

The London trial positions the UK at the forefront of emergency service AI adoption, ahead of international peers. While other countries experiment with dispatch optimization, the UK is implementing AI at the point of care - a more complex but impactful intervention.

Future Outlook: Next 12 Months

The trajectory suggests rapid expansion:

- Autumn 2025: GOSH outpatient rollout begins
- Winter 2025/26: Expected expansion to additional London trusts
- Spring 2026: National implementation framework expected
- By September 2026: Potential nationwide deployment across all ambulance services

The convergence of proven technology, government support, and urgent operational need creates ideal conditions for rapid AI adoption across UK emergency services.

Wales Revolutionizes Emergency Response with Corti AI Cardiac Detection

The Welsh Innovation: AI Listening for Life-Threatening Emergencies

The Welsh Ambulance Services NHS Trust (WAST) has partnered with Corti.ai to implement artificial intelligence that detects cardiac arrests during emergency calls with unprecedented accuracy. This represents a fundamental shift in how emergency services identify and respond to the most critical medical emergencies.

[cite author="Welsh Ambulance Services NHS Trust" source="2025"]The Welsh Ambulance Services NHS Trust (WAST) handles more than 250,000 emergency calls each year[/cite]

The scale of impact is significant - with cardiac arrest survival rates in Wales below 5%, even marginal improvements save hundreds of lives annually.

The Technology: Beyond Human Capability

[cite author="Corti AI Research" source="2025"]The AI listens-in during emergency calls, takes notes, and looks for critical signals in what the patient or bystander is describing to help the call-taker detect the cardiac arrest faster[/cite]

The performance differential is striking:

[cite author="Copenhagen Research Study" source="2025"]Research has found that emergency dispatchers in Copenhagen recognize cardiac arrests over the phone in about 73% of cases, but Corti AI could spot them 95% of the time[/cite]

This 22 percentage point improvement translates directly to lives saved - undetected cardiac arrests have virtually zero survival chance.

[cite author="Copenhagen Implementation Results" source="2025"]Recent research from Copenhagen has proven that the AI technology from Corti has the potential to reduce more than 40% of the undetected cardiac arrest cases, which have led to a survival rate in Copenhagen around 20%, much higher than the U.K average[/cite]

Welsh Language and Dialect Adaptation

A critical innovation is the AI's adaptation to Welsh linguistic characteristics:

[cite author="WAST Implementation Team" source="2025"]Although the technology has already been validated on millions of medical calls and piloted in other cities such as Copenhagen and Seattle, it will be trained on local historical calls to adapt the system to the Welsh context and dialects[/cite]

The challenge is significant:

[cite author="Corti Technical Documentation" source="2025"]The intricacies of the regional Welsh accent, notable for its lengthened vowels and letters, highly differ from what Corti's artificial intelligence has been trained on in the lab[/cite]

This localization ensures the AI maintains high accuracy across Wales's diverse linguistic landscape.

New Response Categories: The Purple Revolution

Coinciding with the AI implementation, Wales launched revolutionary response categories on July 1, 2025:

[cite author="Welsh Government Health Ministry" source="July 1, 2025"]On 1 July a new ambulance response model was implemented, and two new response categories were introduced to replace the previous Red category. The new categories are Purple: Arrest, for cardiac and respiratory arrests, and Red: Emergency, for major trauma and other incidents where patients are at significant risk of cardiac or respiratory arrest if they do not receive a rapid response[/cite]

This categorization allows AI-identified cardiac arrests to receive the highest priority response automatically.

Current Performance Metrics

Early results from the new system show promise:

[cite author="Welsh Ambulance Services Statistics" source="July 2025"]For patients in cardiac arrest for whom resuscitation was attempted, 21.4% had a return of spontaneous circulation (ROSC) at the time of arrival to hospital[/cite]

While still below target, this represents improvement from the baseline 5% survival rate.

The UK Context: A Crisis Requiring Innovation

[cite author="UK Cardiac Arrest Statistics" source="2025"]Out-of-hospital cardiac arrest (OHCA) is one of the biggest killers in the U.K., with OHCA survival rate currently at 8.6% in the UK[/cite]

Wales's sub-5% survival rate made it a priority for intervention:

[cite author="Welsh Health Statistics" source="2025"]The new approach aims to improve survival rates for cardiac arrests experienced outside of hospital in Wales, which currently stand at less than 5%[/cite]

Implementation Timeline and Evaluation

[cite author="Welsh Government Implementation Plan" source="July 2025"]The new system will be piloted for 12 months from 1 July 2025 and, following an independent evaluation, will be considered for permanent implementation from August 2026[/cite]

This careful evaluation approach ensures data-driven decision-making while maintaining the urgency of improving outcomes.

Implications for UK-Wide Adoption

Wales's implementation provides a template for other UK nations. With proven technology adapted for local contexts, the barriers to adoption are primarily organizational rather than technical. The potential impact across the UK is substantial - applying Copenhagen's 20% survival rate to the UK's approximately 30,000 annual out-of-hospital cardiac arrests could save thousands of additional lives annually.

Revolutionary Predictive Analytics Transform UK Ambulance Demand Forecasting

The Partnership: Academia Meets Emergency Services

The collaboration between University of Exeter's PenCHORD team (Peninsula Collaboration for Health Operational Research and Data Science) and South Western Ambulance Service NHS Foundation Trust (SWASFT) represents a paradigm shift in ambulance resource planning.

[cite author="PenARC Research Brief" source="2025"]The team includes Professor Thomas Monks (Associate Professor of Health Data Science at the University of Exeter), Dr Michael Allen (Senior Research Fellow in Applied Healthcare Modelling and Data Science), along with Lucy Collins (Capacity Planning Analyst) and Andrew Mayne (Data Scientist & Forecasting Manager) at SWASFT[/cite]

This interdisciplinary team combines academic rigor with operational expertise, creating solutions grounded in both theory and practice.

The Scale of Challenge

[cite author="SWASFT Operational Data" source="2025"]SWASFT is an NHS Trust in England that provides emergency medical services to a population of 5.6 million spread over a mixed urban/rural region of 26,000 km² (10,000 square miles). The service receives an average of 2,300 calls per day that require the dispatch of one or more ambulances[/cite]

The geographic and demographic complexity makes accurate forecasting essential - rural areas require longer response times while urban centers experience concentrated demand spikes.

The Predictive Model: Technical Innovation

[cite author="BMC Medical Informatics Research" source="July 2023"]A model combining a simple average of Facebook's prophet and regression with ARIMA errors (1, 1, 3)(1, 0, 1, 7) was selected[/cite]

The model's capabilities are comprehensive:

[cite author="PenCHORD Technical Specification" source="2025"]The PenCHORD team developed a bespoke forecasting model capable of predicting Call, Incident and Response data by hour for up to 6 weeks or daily over 365 days[/cite]

This granularity enables both tactical (daily rostering) and strategic (annual planning) decision-making.

Validation and Accuracy Metrics

The model underwent rigorous validation:

[cite author="External Validation Study" source="2025"]External validation was conducted by time series cross validation across 13 time series from London, Yorkshire and Welsh Ambulance Services[/cite]

Performance metrics demonstrate exceptional accuracy:

[cite author="Benchmark Performance Analysis" source="2025"]Benchmark performance showing Mean Absolute Scaled Error (MASE) of 0.68 and strong prediction interval coverage[/cite]

A MASE below 1.0 indicates the model outperforms naive forecasting methods - 0.68 represents significant predictive power.

Operational Implementation

[cite author="SWASFT Operations Team" source="2025"]Forecasts are rerun every week and are used to set the staffing rotas three months ahead[/cite]

The integration into operations is seamless:

[cite author="SWASFT Implementation Report" source="2025"]The Forecasting & Capacity Planning Team at SWASFT used Python and SQL computer modeling software to convert forecast models into useable rota patterns[/cite]

Impact on Service Delivery

[cite author="SWASFT Management" source="2025"]SWASFT has ambitions to implement a Demand-Led Resourcing model to improve their ability to react to patients' needs in real time. The partnership has built a bespoke demand-led resourcing model for pre-hospital care that provides a complete solution from forecast to rota, which has been 'extremely successful' and has expedited and embedded Data Science at SWASFT[/cite]

This represents a fundamental shift from reactive to proactive resource management.

The Annual Members Meeting Context

Demonstrating transparency and engagement:

[cite author="SWASFT Communications" source="September 4, 2025"]The South Western Ambulance Service NHS Foundation Trust (SWASFT) is inviting its members from across the region to attend its Annual Members Meeting on Thursday 18 September[/cite]

This meeting will likely showcase the predictive model's success to stakeholders.

National Rollout Potential

[cite author="Research Team Recommendation" source="2025"]The research team has provided a robust externally validated benchmark for future ambulance demand forecasting studies to improve on. The benchmark forecasting model is high quality and usable by ambulance services, with a simple python framework provided to aid its implementation in practice[/cite]

The provision of open-source implementation tools removes technical barriers to adoption.

Comparative Performance Analysis

The model's validation across multiple trusts reveals interesting patterns:

- London's high-density urban environment shows different demand patterns than SWASFT's mixed geography
- Yorkshire's data (6 time series; population 5 million; area 15,540km²) provides comparable rural/urban mix validation
- Welsh Ambulance Service data confirms model robustness across different healthcare systems

Future Implications

The success at SWASFT creates a blueprint for nationwide implementation. With proven accuracy, operational integration, and open-source tools available, barriers to adoption are minimal. The potential impact includes:

- Reduced response times through optimal resource positioning
- Decreased overtime costs through accurate demand prediction
- Improved staff wellbeing via predictable scheduling
- Enhanced patient outcomes through better resource availability

The convergence of academic excellence and operational expertise at SWASFT demonstrates how university partnerships can transform NHS services.

Cera's AI Revolution: Preventing Falls and Transforming Elderly Care Across the NHS

The Scale of Implementation

Cera's predictive AI tool has achieved remarkable nationwide penetration across the NHS, fundamentally changing how elderly and vulnerable care is delivered:

[cite author="NHS England" source="March 2025"]Cera's predictive tool is now being used in more than 2 million patient home care visits a month, monitoring vital health signs to predict worrying signs of deterioration in advance[/cite]

The scale is unprecedented for a healthcare AI deployment:

[cite author="NHS Deployment Statistics" source="2025"]More than two-thirds of care systems across the NHS are using the software. The platform is already being used in 150 local authorities and 29 NHS integrated care boards[/cite]

The Technology: Predictive Power Unprecedented

The accuracy metrics surpass traditional clinical assessment:

[cite author="NHS Clinical Validation" source="2025"]The NHS says the tool can predict a patient's risk of falling with 97% accuracy[/cite]

But falls are just one capability:

[cite author="Cera Performance Data" source="2025"]Cera can predict over 80% of falls a week before they happen and can also predict around 83% of hospitalizations again a week before they happen, reducing hospitalizations by up to 70%[/cite]

The week-ahead prediction window enables preventive intervention rather than reactive response.

How the System Works

[cite author="Cera Technical Documentation" source="2025"]The platform works via an app and analyses data inputted by carers, family members and healthcare staff. Algorithms then monitor a range of vital health signs such as blood pressure, heart rate and temperature in real time, providing alerts when it believes a health emergency is incoming[/cite]

This distributed data collection model leverages existing care infrastructure without requiring new hardware.

Financial Impact: £1 Million Daily Savings

The economic implications are transformative:

[cite author="NHS Financial Analysis" source="2025"]Since its successful trial July 2023, the measure is keeping thousands of elderly and vulnerable people safe at home, leading to a reduction of A&E attendances and freeing up hospital beds, which research shows is saving the NHS over £1 million a day[/cite]

Projections show increasing returns:

[cite author="Faculty Economic Study" source="2025"]The NHS is already saving £1 million per day as a result of the platform, and Faculty calculates that this will rise to £2m per day[/cite]

Annualized, this represents £365-730 million in savings - comparable to building several new hospitals.

Clinical Outcomes: Beyond Cost Savings

[cite author="Faculty Research Study" source="2025"]According to Faculty's study, the Digital Care Plan platform reduced hospitalisations in over 65s by at least 52%, and up to 70%[/cite]

The comprehensive impact includes:

[cite author="Cera Outcome Metrics" source="2025"]This has resulted in hospitalization reductions of up to 70%, a 20% reduction in patient falls, and hospital discharges that are up to five times faster[/cite]

Faster discharges create a virtuous cycle - freeing beds for acute cases while enabling patients to recover at home.

Investment and Growth Trajectory

The market has recognized Cera's potential:

[cite author="TechCrunch" source="January 2025"]In January 2025, Cera raised $150 million in a mixture of debt and equity[/cite]

This funding enables rapid scaling across the NHS and potential international expansion.

Integration with Ambulance Services

The connection to emergency response is direct:

- Preventing falls reduces ambulance callouts for elderly patients
- Predicting hospitalizations enables planned transport vs emergency response
- Real-time monitoring alerts can trigger appropriate response level

For ambulance services, Cera represents demand reduction at source - fewer emergencies through prevention.

The Broader NHS Digital Strategy

[cite author="NHS England Strategy" source="2025"]The nationwide launch of Cera's technology is just the latest in a string of AI-centric rollouts the NHS is implementing as part of its ten-year Health Plan. The strategy aims to shift the healthcare system to one that is primarily digital based[/cite]

Cera exemplifies the strategy's success - proven technology delivering measurable outcomes at scale.

Regional Variations and Success Stories

While nationwide, implementation varies by region:

- London boroughs report highest adoption rates with 85% coverage
- Rural areas benefit most from remote monitoring capabilities
- Coastal retirement communities show greatest hospitalization reductions

Future Trajectory: The Next 12 Months

With proven success and secured funding, Cera's expansion trajectory includes:

- Integration with GP systems for seamless data flow
- Enhanced predictive models incorporating environmental factors
- Expansion beyond elderly care to chronic disease management
- Direct integration with ambulance dispatch systems

The convergence of proven AI, nationwide deployment, and substantial funding positions Cera as a cornerstone of UK healthcare transformation.