Back

Case study

eConsult’s AI transcriber frees up NHS GP time

GPs and other NHS healthcare clinicians working in the NHS face huge time pressures to do their jobs effectively. A factor that makes this pressure worse is the time it takes the clinicians to write up notes from patient consultations and devise treatment plans – with documentation taking up to a third of their precious time.  

AI is helping to reduce this pain point in the form of a new transcription summariser tools from health tech innovator eConsult – which is now part of global healthcare business Huma. Its new generative AI-based documentation automates and streamlines the clinical note taking process. Opencast has helped eConsult to take its idea from concept to reality, and make it distinctive from its competitors. 

A small Opencast developer and quality assurance (QA) team worked with eConsult on the creation of ‘Hi Scribe’. The new tool is secure, easy to use, enhances accuracy and gives GPs more time to focus on what really matters – which is their patients.  

Huma launched the new tool in May 2025. Watch their video explaining how Hi Scribe works. 

  • 33% less time

    Total clinician time to take patient notes 

  • From 5 minutes to a few seconds

    Time taken to complete notes 

  • Multiple integration

    With existing NHS systems 

Download case study

Fill out the short form and we’ll email you a link to download a PDF of the case study.

*
*
*

Challenge

Cut clinician time for consultation note-taking

Healthcare clinicians working in and beyond the NHS are under enormous time pressures to do their jobs – and a key pain point is the time it takes them to write up the notes from patient consultations. Up to a third of clinician time can be taken up with documenting consultations before coming up with recommendations for treatment. This often results in GPs working many hours of overtime. 

The GP founders of health tech scale-up eConsult knew from experience that the time taken for notes and treatment plans was a challenge.  

It saw the potential for AI to help with both of these tasks. It could potentially ease the burden of note taking and offer intelligent suggestions for next steps in treatment. But, to turn this vision into reality, they needed specialist support. 

Any technology response would have to comply with the stringent regulation of data and patient protection. Patients would also need to be confident that any use of technology deployed would not compromise doctor-patient confidentiality.  

Clinicians would need to be able to use any technology easily, including being able to integrate it seamlessly into electronic patient records.  

Importantly for eConsult/Huma, the new AI product would also need to work commercially. The business would need certainty that their vision was achievable and that it would fit in a marketplace where it would not be the only player. 

The project would help eConsult/Huma move from a purely conceptual stage with a back-end API to a viable solution with a user-facing front end. 

Response

A regulation-compliant and secure AI tool

Responding to the challenge, eConsult’s first step was to identify more about what clinicians needed to ease time pressures. Opencast’s developer and QA capabilities helped eConsult to define user needs and understand the competition through market and user research. The immediate aim on that back of that was to create a minimum viable product (MVP) for the new AI tool.  

Once through discovery, Opencast developer expertise moved to design and build the new tool, using lean Agile processes. Prototyping and spiking technology, which explores potential solutions, helped the team identify viable and cost-effective solutions that could be integrated into the platform.  

Constant iteration and feedback loops informed next steps, consulting on and defining which features the new tool really needed and which were ‘nice to have’. What was definitely essential throughout was the need to devise a product that would be secure and compliant with government regulations and NHS data and information standards, work seamlessly with other NHS technology – and be competitive in a fast-growing marketplace. 

With a small team in place, Opencast’s approach to the project was characterised by pragmatism throughout, with QA and developer roles closely intertwined and the team staying flexible enough to take on tasks across both disciplines.  

The team modernised testing practices, with a strategy that allowed the team to find and resolve issues faster, improving overall efficiency and the robustness of the eventual Hi Scribe solution.  

A key aim of the project was balancing the use of AI to deliver core functionality at the same time as ensuring data protection for patients.  

The Opencast QA on the project integrated advanced AI and machine learning services for functionality. To maintain data privacy and compliance, AWS Comprehend was deployed for natural language processing, importantly including its function for identifying and redacting personally identifiable data. 

Moving fast, in the first four months of 2025, the team was able take the eConsult concept to private beta then into a potentially saleable product in public beta.  

In May 2025, Huma, the global healthcare AI company that had acquired eConsult only seven months earlier to accelerate its drive for digital-first healthcare, publicly announced the launch of Hi Scribe. 

The new Hi Scribe product is built on top of Google’s Vertex AI platform, which Huma uses for its Huma AI foundation engine to power the company’s other AI-first tools and products. It leverages Huma’s strategic partnership with Google Cloud to deliver secure, scalable and clinically compliant AI services to healthcare workers.  

Data is the fuel for the Hi Scribe product – and it is powered by insights from over 60 million consultations conducted by eConsult. 

The new Hi Scribe tool transcribes patient-clinician consultations in real time and automatically generates structured clinical notes for direct integration into NHS and other electronic health record (EHR) systems. This integration enables clinicians to submit documentation seamlessly without disrupting workflows.  

The Hi Scribe technology features writeback and coding capabilities for EMIS Health and UK clinical information system SystmOne TPP. And it uses the SNOMED clinical vocabulary, which helps sets Hi Scribe apart from other AI products. 

Compliance is at the core of the new tool. It complies with April 2025 NHS England guidance on the use of AI-enabled ambient scribing products in health and care settings, and with regulations set by the UK Medicines and Healthcare products Regulatory Agency (MHRA) as a medical device.  

"This tool brings clinicians’ attention back to their patients. By leveraging generative AI with strong regulatory governance, we’re ensuring safe, effective and efficient care. We’re excited that, in the near future, on the back of 60 million consultations, we can start helping care providers with intelligent suggestions for next steps – ensuring that nothing is left unnoticed."

Dr Murray Ellender, GP, founder of eConsult and Huma’s Head of UK Healthcare

Technical details

React and TypeScript – to create a responsive, intuitive and user-friendly interface 

Node.js and Express.js (using TypeScript) – to build robust and scalable APIs. 

MongoDB – for data management

Auth0 – for secure authentication

Advanced AI and machine learning services – for core functionality and data privacy 

AWS Comprehend – for natural language processing and to redact personally identifiable data for privacy and compliance 

CookieFirst – to manage user consent and ensure compliance 

Playwright tests – integrated into CI/CD pipeline, migrating from periodic

Impact

A time-saving, user-friendly tool that GPs like

With the system just into public beta in early summer of 2025, it was early days to prove the full impact of the Hi Scribe tool. But time saving was easily the best and most impactful change of the new product.  

Where a GP would typically take three to five minutes to write notes and consider treatment, Hi Scribe can help them to generate a consultation summary and write this into the notes in a matter of seconds.  

The fully functional front-end application for Hi Scribe was tested with a select group of GPs, who offered positive feedback that highlighted its ease of use and time-saving potential. The GPs highlighted the user-friendliness of the Hi Scribe front-end. 

One London-based GP, Ross Dyer-Smith, said: “Using Hi Scribe to transcribe my medical notes has transformed my consultations – saving time, reducing pressure and letting me focus fully on my patients. I have more headspace to think, listen, and make decisions. It’s brought back more of the most satisfying part of practising medicine – truly engaging and connecting with my patients.” 

This positive feedback from GPs was supported by a detailed market research and competitor analysis report.  

Opencast teams always work to foster collaboration and knowledge transfer to and from its partners. The Hi Scribe project was no exception.  

Our developer joined an engineering weekly demo to talk about what had been achieved and shared progress with other teams. And our QA led community-of-practice sessions for the wider eConsult QA team, offering a platform for sharing best practice, discussing challenges and exploring new techniques in quality assurance. This contributed to overall skills development and collaboration within the in-house QA team. 

Strategically, Hi Scribe has delivered an entirely new type of service, capable of integrating with key clinical systems like EMIS and SystmOne, significantly expanding its potential reach and utility within the healthcare ecosystem.  

Huma's 2024 acquisition of eConsult has further amplified the project's value. Hi Scribe is now recognised for its potential to play a pivotal role in Huma's strategic Huma Intelligence initiative, which aims to embed more AI into healthcare tools, showcasing the project's lasting and evolving impact. 

Murray Ellender, the GP founder of eConsult who is now Huma’s Head of UK Healthcare, said: “This tool brings clinicians’ attention back to their patients. By leveraging generative AI with strong regulatory governance, we’re ensuring safe, effective and efficient care. We’re excited that, in the near future, on the back of 60 million consultations, we can start helping care providers with intelligent suggestions for next steps – ensuring that nothing is left unnoticed.” 

  • Secure data solution

  • User friendly

  • Enhanced accuracy

  • Commercially viable

Related case studies

Loading...