Case Study
Ambient Voice Technology transforming healthcare for patients and clinicians across London
CHILD HEALTH AI NETWORK (CHAIN) THEME
Ambient Voice Technology rolled out across 9 sites in London increased direct patient care and capacity, improved patient-clinician interactions, and could save 47 minutes in an Emergency Care shift.
Unmet Need
While the electronic health record (EHR) brings clinical records together in one place, it also creates an administrative burden on patient-facing clinical staff who need to spend time documenting important conversations and patient data during and after appointments. This contributes to clinical staff burnout, poor patient-clinician interaction during visits, and patient-facing staff spending more time on data entry.
Solution
TORTUS, an AI documentation platform, uses Ambient Voice Technology (AVT) to streamline workflows, reduce EHR interaction time, and maximise direct patient care by efficiently capturing dialogue during appointments and organising clinical data into notes. The AVT technology was developed and successfully rolled out across multiple sites in London in a range of clinical contexts including paediatrics and emergency care.
Support Provided by the HRC
The Great Ormond Street Hospital Data Research, Innovation and Virtual Environments (GOSH DRIVE) Unit, the team behind our Child Health AI Network (CHAIN) Theme, led the development, rollout and evaluation phases of the technology.
Funding from the NIHR HRC PCH provided two staff members who managed the data collection and offered clinical and operative guidance. The collaborative partnership between GOSH and NIHR also influenced the decision to initiate the work, as it aligns with the objective to establish the NIHR HRC PCH as a lead centre for artificial intelligence and machine learning tool testing in collaboration with industry.
Project Progress
FROM
TRL STAGE
TO
TRL STAGE
Outcomes & Impact
PRODUCTIVITY AND SERVICE IMPROVEMENT
The use of AVT improved clinician satisfaction with time, attention and documentation accuracy. It enabled clinicians to focus on patients rather than screens and resulted in significant reduction in stress and mental effort (as measured by the NASA Task Load index). Appointments were 8% shorter and a higher percentage of patient notes were completed after a session with AVT use.
Evaluation at the Emergency Department demonstrated a 13.4% increase in patient capacity per shift and a 51.7% reduction in documentation time, which would save an average of 47 minutes per shift. This increased efficiency would enable clinical staff to see one additional patient per shift, which would amount to 9,259 additional A&E attendances per day if scaled nationally.
PATIENT BENEFIT
Overall, key findings include a 23.5% increase in direct care with less time spent on administrative tasks, and improved patient-clinician interactions.
Parents/carers reported a more positive experience with AVT and were likely to recommend its use in a clinic to a family member/friend.
Some patients reflected that they felt the benefit of receiving their doctor’s full attention, felt more able to stay focused in the appointment, and that the tool wasn’t intrusive in their appointments.
FUTURE INNOVATION PRACTICE
The structured methodology used to deploy and evaluate AVT across multiple sites led to the inception of the NHS T.E.S.T framework, a rigorous model to assess the safety, efficacy, and scalability of a technology. This has been recorded in a playbook, available for any innovative partners working in, ow with, the NHS.
The final report of the project is available and includes more information on the playbook (p. 92) and NHS T.E.S.T (p. 101).
This project forms part of the GOSH AI strategy, which aims to deliver real world AI solutions that improve patient outcomes, our corporate functions and the working lives of our staff in a safe and sustainable way.
PUBLICATION AND DISSEMINATION
Several publications have resulted from the work behind this project.
The report of this evaluation has been publicised in several national newspapers, including The Independent and Evening Standard, and was explicitly mentioned in the UK government’s “10-year plan” as it outlines intention to implement Artificial Intelligence across health and care.
This project has been presented at several conferences including the Child Health Technology Conference and the Great Ormond Street Hospital Conference in 2025.
Quotes
“Very good appointment today and I like the new approach. It is easier to have a speedy appointment due to my child’s disability”.
Parent/Carer
Partners
