We’re delighted to have you joining us at the School to take up the position of Professor of Interventional Image Computing. Could you start by telling us a little bit more about your research?

My ambition is to accelerate the clinical translation of innovative interventional imaging algorithms and devices for a range of medical applications. Within the interventional engineering and surgery domain, machine learning is something I am really striving to develop. Classically, interventional imaging devices were simple and mostly restricted to providing direct imaging output. As the field progresses, surgeons have an increasing number of tools and features integrated in the operating theatre which is significantly increasing the amount of information available. Having access to richer information is of course desirable but leads to a substantial cognitive load for the clinical team in situations where clear decisions and actions are of paramount importance. Machine learning can play a crucial role to provide solutions for our clinical collaborators by exploiting, linking and summarising the vast amount of data generated by the multiple information streams.  

I’m also really keen to push best practice and collaboration within the sector. We recently launched the NiftyNet research consortium which is the first open-source deep learning software library dedicated to medical imaging. With this tool we can support the translational aims of the whole research community by pooling resources within the field to develop more robust algorithms, which in turn will benefit patients.

What do you think is the biggest challenge facing surgical sciences?

Our community has been producing beautiful research solutions which, unfortunately, have seldom been picked up by our clinical collaborators. If clinical translation does not happen at the research level, there’s even less chance of the solutions being adopted by industry. This means the work often ends up as a nice paper but is not progressed any further. I feel this challenge is often down to a disconnect between the research and industry mindsets. On the one hand, research groups feel they have demonstrated the science, so it should therefore be suitable for translation. On the other hand, industry are often very risk adverse in taking new research ideas forward into products, particularly if a concept isn’t mature enough. This creates what we call the ‘translational gap’.

I’m really keen to work towards bridging this disconnect by finding better ways to demonstrate the usefulness of our research and ensuring we are always considering feasibility, robustness and safety standards, the question of which instantly becomes critical when we are talking about surgical workflows. The infrastructure and resources needed to bridge this gap are significant, as the initial return on investment for research outcomes are not well balanced and needs to be much higher than you would expect for a standard medical imaging research piece. However, the pathway to impact that this will eventually create will make that initial investment worthwhile for patients and healthcare systems.

What would you predict will be the next ‘big innovation’ to change clinical practice in surgery?

I think it won’t be long until we begin to see really personalised decision support tools which use machine learning to exploit the wealth of data from population studies and clinical trials to support everything from surgical planning and assessment through to delivery and outcomes. This would feed into the next generation of ‘smart’ devices which can adapt to their specific environment; be that the type of surgical application, patient-specific considerations or both combined.

What has been one of the most interesting research projects you’ve worked on during your career?

Since 2014 I have been a co-investigator on the GIFT-Surg project (Guided Instrumentation for Fetal Therapy and Surgery). In fact it’s actually what brought my career back into academia. When I initially heard about the project I thought it was science fiction and didn’t even know surgery in utero was possible. It’s quite a niche area but also one where the potential impact for patients and families is huge. It’s also an area where the constraints of the surgery, the need for efficacy and the ethical considerations are particularly stringent as compared to any other surgery I can think of. It’s definitely a uniquely interesting and complex research challenge to be able intervene in such an environment. It forces the research team to think out of the box.

You mention your move from industry to academia, what have been the biggest differences between the two sectors for you personally?

What I really like with research in industry is that you see the impact of what you’re doing almost immediately. The time between research and subsequent impact on patients is relatively rather short; perhaps a couple of months when working on a small improvement before a product release or a couple of years for more substantial pieces. It’s very rare you would work on a project for the sake of doing research, there is always product development in mind. That drive to always improve existing solutions with a view to creating impact for patients is something I don’t see as much in academia, and that’s where I think I can contribute. On the flipside, in academia you have relative freedom to explore concepts and avenues which might end up producing a serendipitous discovery. There is also the potential to find new synergies across clinical applications - in industry it’s unlikely that will happen due to the focused remit of the tools you would be working on. However, what I love the most about academia is the chance to work with early career researchers who have a real ambition to change things, it’s very rewarding to mentor them and help them to develop as researchers.

The way funds are generated in academia must have also been a change, what are your thoughts on the funding landscape within healthcare engineering?

Having large non-governmental funding bodies in the UK such as Wellcome is quite unique. It’s not something I’ve seen in France or even the US. There are some very nice alignments on innovation strategy with the major funding bodies here. I think the UK is extremely well positioned to address the translation gap and we should be taking advantage of this in our research community.

You hold a guest professor position with the Organ System Unit at KU Leuven, could you tell us a little bit about this collaboration?

As we work at the interface between engineering and surgery I think it’s critical we don’t limit ourselves to just working with an engineering faculty. This appointment is to support my work with the aforementioned GIFT-Surg project in particular as part of which I began to collaborate closely with Prof Jan Deprest, who leads the Organ System Unit in KU Leuven.

To help us ensure translational success, our engineering PhDs always have a clinical co-supervisor who will be able to mentor their research. In return, I act as an engineering link to provide technical advice for the clinical fellows and students which our clinical collaborators such as Jan supervise. Additionaly, we observe strong direct interactions between the engineering PhD students and the clinical fellows. This creates a feedback loop to ensure clinical reality is built into the engineering developments we are working on. In addition, although I’ve never lived in the country, the position also offers a nice connection to my historical roots and Belgian nationality. 

Finally, anything you would like to mention to our readers? 

The School’s position within a medical faculty is one of my main motivations to join King’s. I believe it will really help close those gaps we’ve discussed and try and ensure we can translate our research as efficiently as possible.

I’m already excited by the very warm welcome I’ve received so far and the supportive atmosphere. I really look forward to building new collaborations with the team here and taking advantage of the strong interaction between the university and the NHS trusts.