Q-BIOMED Knowledge Exchange Series: What our regulators are really telling us
On 20 April 2026, Q-BIOMED convened its first formal Regulatory Knowledge Exchange, bringing together representatives from NICE, MHRA, BSI and ABHI for an unusually candid conversation with clinicians, physicists, engineers and researchers working at the frontier of quantum-enabled healthcare.
There were no submissions under review. No approvals pending. No polished pitches. Instead, the session created something far rarer: a space for honest dialogue about the distance between breakthrough science and clinical reality, and the relationships, language and evidence required to bridge it.
Who was in the room?
For the first time, the Q-BIOMED Clinical Steering Group brought together four major regulatory and standards bodies, in person and online, for an unscripted exchange with the UK quantum sensing community. The significance of that gathering was not procedural but cultural.
Too often, researchers encounter regulation only at the point of translation, when timelines tighten, investment pressures mount and fundamental questions about evidence, classification and clinical utility suddenly become unavoidable. This session inverted that model entirely. The purpose was not to seek permission, but to cultivate understanding early enough for it to meaningfully shape innovation.
Again and again throughout the discussion, the same theme surfaced: regulation works best not as a late-stage hurdle, but as a conversation woven into the research process from the beginning.
Where does the regulatory journey actually begin?
For many researchers, regulation appears as a distant finish line: a CE mark, a UKCA mark, a reimbursement decision somewhere far beyond the laboratory.
Steve Lee (ABHI) immediately challenged that framing. CE or UKCA marking, he explained, is not the beginning of the journey. It is the culmination of it. The real starting point lies much earlier, in a handful of deceptively simple questions:
Does the technology meet the legal definition of a medical device?
What is its intended clinical use?
Who exactly is the intended user?
And if clinical studies are planned, will they constitute formal clinical investigations requiring MHRA authorisation, or broader exploratory research outside device regulation?
Dr Andrew Zambanini (MHRA) pushed the discussion even further upstream. The question he hears too rarely, and often far too late, is not what can this technology do, but what is it actually for? This is not what the physics allows, what the sensor can theoretically measure. Instead, this relates to what problem in clinical practice it will genuinely solve, replace or improve.
The point was not pessimism, but precision. A great deal of this strategic thinking can happen long before a device exists physically. Early modelling, evidence planning and pathway mapping can all be done on paper or in silico. What matters is ensuring that the research questions being asked today are already aligned with the evidence regulators, clinicians and payers will eventually require tomorrow.
Clinical investigation or clinical research?
One of the session’s most practically useful clarifications concerned the distinction between exploratory clinical research and a regulated clinical investigation.
Steve Lee explained that the dividing line is intent, if a study is designed to establish the safety or performance of a medical device, it constitutes a clinical investigation and requires MHRA authorisation before commencement. If instead the work is exploratory, investigating biology, mechanism or feasibility without claiming diagnostic or performance outcomes, it generally sits outside that framework.
Professor Pier Lambiase illustrated the distinction elegantly. Studies using hyperpolarised MRI to explore glucose metabolism in tumours, for example, remain research-focused investigations into biological behaviour. By contrast, studies establishing diagnostic sensitivity, specificity or clinical decision thresholds move decisively into regulated territory. Both kinds of work are scientifically valuable, but only one initiates the formal regulatory pathway.
Where NICE enters the picture
Dr Anastasia Chalkidou (NICE) described a model of evaluation that surprised many in the room. Contrary to common assumption, NICE does not engage only once a technology is mature and fully evidenced. Its Early Value Assessment (EVA) programme exists specifically to support promising technologies whose evidence base is still evolving.
A CE or UKCA mark remains a prerequisite for entry into EVA. Safety and conformity must already be established. But the evidentiary threshold at that stage is intentionally lighter. Rather than issuing a definitive reimbursement decision, NICE can provide a conditional recommendation while identifying the two or three outcomes most critical for future evaluation. Importantly, NIHR funding may then support the generation of that targeted evidence through competitive schemes aligned with EVA recommendations.
The broader message from NICE was one of integration. Treating regulatory approval and health technology assessment as separate, sequential exercises is one of the most expensive mistakes innovators make. The evidence required for MHRA and NICE frequently overlaps, and studies designed with both regulatory and reimbursement objectives in mind can dramatically reduce duplication later.
This joined-up approach is already visible in initiatives such as the Innovative Devices Access Pathway (IDAP), developed collaboratively between MHRA and NICE and currently under consideration for broader expansion.
OPM-MEG for Epilepsy: A case study in translation
The discussion became particularly animated when Dr Umesh Vivekananda introduced OPM-MEG (optically pumped magnetoencephalography) as a live case study. Suddenly, abstract regulatory principles acquired texture and concrete technological questions prompted concrete regulatory responses.
The clinical context
Epilepsy diagnosis has relied largely on EEG for nearly a century and on structural MRI for around fifty years. Standard thirty-minute EEG recordings detect epileptiform activity in only around 35–50% of cases. Conventional MEG improves detection rates substantially and has become valuable for pre-surgical localisation in refractory epilepsy. Yet conventional MEG systems remain prohibitively expensive, requiring cryogenic cooling infrastructure and existing in only a single active clinical centre in the UK.
OPMs offer a striking alternative by replacing cryogenic sensors with compact rubidium or helium vapour-cell sensors operating at room temperature, OPM-MEG systems can be mounted within wearable caps that allow natural patient movement during recording. The UCL team, working with Mag4Health in France, is developing the technology initially for epilepsy surgery planning and source localisation, with longer-term ambitions extending toward earlier diagnosis and prolonged telemetry studies.
What the regulators asked
Interestingly, Andrew’s first questions were not clinical but commercial. Who owns the intellectual property? Who is developing which components? What is the route to market?
These questions reflected a central regulatory reality: MHRA regulates products that ultimately reach patients through real-world clinical systems. Umesh Vivekananda explained that UCL intends to retain ownership of the software, clinical interface and wearable hardware architecture, while preserving long-term sensor flexibility across manufacturers.
Andrew’s advice was pragmatic and memorable. Attempting to certify a fully AI-enabled diagnostic platform from the outset risks multiplying regulatory complexity before the underlying clinical pathway is established. Put simply, get the device into expert clinical use first. Then generate meaningful real-world data and add AI functionality later.
Several additional themes emerged:
AI-driven analysis pipelines are likely to fall under Software as a Medical Device (SaMD) regulation.
Paediatric AI training datasets remain particularly challenging for emerging modalities such as OPM-MEG.
MHRA’s AI Airlock initiative is expanding, though access remains competitive.
NICE’s Evidence Standards Framework already provides a route for assessing digital and AI-enabled technologies.
Health economic modelling can begin surprisingly early, particularly when technologies offer extended monitoring advantages over existing standards of care.
The exchange offered a vivid reminder that translational success depends not only on technical excellence, but on sequencing: understanding which questions must be answered first, and which layers of complexity can be introduced later.
What happened after the session
Following the meeting, both Andrew and Steve shared further written reflections addressing questions the discussion had not fully explored on the day. Several themes proved particularly significant.
1. Standards must be international
Steve emphasised the importance of developing standards that are internationally aligned rather than narrowly UK-specific. This resonated with Ben Sheridan (BSI), who reflected that as the UK National Standards Body, the BSI approach is to create international-first standards. For UK innovators hoping to scale globally, fragmented standards frameworks create friction, duplication and delay.
The standards community is therefore actively exploring new forms of pre-regulatory engagement and sandboxing mechanisms designed to support innovators earlier in development, before formal regulatory pathways begin.
2. CE Marking and UKCA: the transitional reality
The session also clarified the evolving relationship between CE marking and UKCA requirements. Northern Ireland remains aligned with EU MDR and IVDR frameworks, meaning UKCA marking alone is insufficient there.
Within Great Britain, however, CE-marked devices continue to be recognised under transitional arrangements extending through various dates between 2028 and 2030 depending on device classification. Most notably, MHRA launched a February 2026 consultation proposing indefinite recognition of CE-marked devices within Great Britain. A final policy decision has not yet been announced.
3. Quantum is not a separate regulatory category
Perhaps the most surprising revelation for some attendees was Andrew’s observation that MHRA does not currently treat quantum technologies as a standalone regulatory domain. Quantum sensing and quantum computing are not regulated because they are quantum.
The regulatory pathway follows the clinical purpose rather than the underlying physics. What matters is the clinical function of the resulting product. Is it a medical device, or an imaging platform, or a diagnostic tool, or Software as a Medical Device?
In many ways, this is reassuring. There is no hidden “quantum-specific” regulatory labyrinth waiting beyond the horizon. Existing frameworks already contain the mechanisms required to evaluate safety, performance and clinical value.
It was agreed that formal mechanisms, including scientific advice, innovation pathways, sandboxes and cross-regulator roundtables, are most effective when used as part of an ongoing conversation, not one-off checkpoints.
The strongest message of the day: come earlier
If one practical lesson echoed consistently throughout the exchange, to engage earlier than you think you need to. Not once a prototype exists, or once clinical trials are being planned, and certainly not once regulatory obstacles emerge.
Scientific advice services become valuable once technologies approach clinical investigation stages, while more specialised regulatory advice pathways exist for later product-specific questions. MHRA’s Innovation Office offers free early-stage guidance intended precisely to help developers identify potential regulatory challenges before they harden into design limitations.
The same principle applies to standards development. Ben encouraged researchers to engage directly with standards committees, contribute to draft guidance and help shape emerging frameworks before they become fixed. For disruptive technologies, standards cannot simply be imposed retrospectively. They must evolve alongside the technologies themselves.
His closing reflection lingered with many participants long after the session ended. Good practice for emerging technologies, he argued, cannot be defined by regulators working in isolation. It has to emerge collaboratively from the scientists, engineers, clinicians and developers who understand both the capabilities and the limitations of the technologies they are building. The standards community does not simply want to assess innovation once it arrives. It wants to help shape the shared language that allows innovation to become clinically trusted in the first place.
Perhaps that, more than anything else discussed that day, was the true purpose of the exchange, not merely understanding regulation, but building the relationships and vocabulary that will eventually allow quantum technologies to move confidently, and credibly, into healthcare.
Participants
-
Programme Director, HealthTech; leading the management and delivery of the HealthTech Programme within NICE’s Centre for Health Technology Evaluation.
-
Emerging Technologies Lead at the National Standards Body, working primarily at the pre-regulatory frontier of standard-setting for novel technologies.
-
Physician and Clinical Pharmacologist; Head of Innovative Products within the Innovation Accelerator. With 25 years of industry experience in drug development, Andrew’s team spans all healthcare products and works in close partnership with MHRA’s medical devices division.
-
Representing the industry body for medical devices, with a membership of 400-500 companies. A regulatory specialist with 30 years of experience, including a formative period at the MHRA, and a background spanning both clinical practice and manufacturing.
We are deeply grateful to all four regulatory guests for their openness, generosity and willingness to engage at such an early stage.
Written by Prof. Sheena Visram, Honorary Professor of Practice and Q-BIOMED Principle Advisor, UCL