29 July 2022
Q&A with Chief Scientist Francesca Cormack
We caught up with our Chief Scientist Francesca Cormack ahead of her AAIC 2022 session on virtual clinical trials. Here Francesca tells us more about her career journey, what she is currently focusing on and why virtual clinical trials are such an exciting area to be working in.
Can you tell us a little bit about yourself and your career journey before joining Cambridge Cognition?
I did my PhD at the University of Newcastle studying visual cognition in dementia, including Alzheimer’s disease, dementia with Lewy Bodies and Parkinson’s disease with dementia. Even at this stage, I was really interested in the use of technology for improving cognitive assessment, making it more efficient, more objective and more reliable. I was coding my own experiments that were delivered on very primitive touchscreen devices that I was carrying around to do testing in the community – they were very heavy!
From there I did my postdoctoral research in paediatric epilepsy, where I learned how to use lots of different tools – not just cognitive testing but also electrophysiology and brain imaging. I then spent some time teaching Master’s courses on research methods and statistics at UCL before moving to the MRC in Cambridge. Here I was developing computerised cognitive assessments to look at different aspects of working memory in both children and adults.
What motivated you to join Cambridge Cognition?
I joined Cambridge Cognition nine years ago. Something that has always cut across my research was an interest in the use of computerised testing to improve our ability to capture cognition across many different populations and demographics. I was also interested in how we can get the most out of all this data we were capturing, using a range of analytic techniques. I’d been familiar with the work of Cambridge Cognition since my PhD and knew they had a reputation for really excellent, neuroscience-based cognitive assessments that had been used across lots of different patient groups. Having gone from disease area to disease area, I knew that the company was a good fit with my interests and strengths, and that I had useful insights to offer from my experience of working across diagnostic groups.
I’d been familiar with the work of Cambridge Cognition since my PhD and knew they had a reputation for really excellent, neuroscience-based cognitive assessments that had been used across lots of different patient groups.
What has your career journey been while working here?
I began as a Senior Scientist, supporting our academic customers and developing analytics for our software as a medical device product. It was a very pleasant surprise to discover the wealth of data that we had and the range of collaborations that were in place.
I then became Director of Research and Innovation, leading the scientific side of R&D. The first project I was involved in was developing cognitive assessments on smartwatches, which was something that nobody had really tried to do before. We were also pulling down the sensor data captured by the watches so we could look at relationships between cognition and physical activity. Around that time, we also started to focus on understanding whether we could implement voice assessments using the cutting-edge automatic speech recognitive technology that was coming out at the time, so this was an exciting time to be working here.
What have been your main focuses since becoming Chief Scientist in April 2022?
Part of my role involves raising the profile of the awesome science that’s done at Cambridge Cognition. We have a lot of talented scientists doing excellent work and one of the really fun bits of my job is getting to go out into the world and talk about it.
We are again at an exciting point in the company. Many collaborations initiated a few years ago are starting to produce a wealth of data, and we are really getting some excellent insights into the value of our tasks across lots of different settings and in combination with other technologies. I really value being part of a great team where we get to think about how best to use these data and what new directions we can go in, from a cognitive, data and technology point of view. We are also very keen to develop new collaborative relationships and to contribute to large-scale science projects and consortia that will not just push us forwards but the field as a whole. It’s great being part of that larger ecosystem.
You are chairing a session on virtual clinical trials at AAIC on 2nd August 2022. What do you find most exciting about virtual clinical trials?
As someone who is quite nerdy about data, one of the things I’m most excited about is the opportunity to collect more data. Not for the sake of it, but to gain information that we just wouldn’t otherwise have. We can assess patients more often which gives us a sense of how they are doing not just at a single time point but how their cognition or their symptoms change on a day-to-day basis. Having this temporal resolution for our measurements was something we didn’t really have until now.
It also gives us the opportunity to combine different data sources. We can capture cognition alongside measures of mood, activity, and sleep. Having that context really is the key to improving the accuracy of active assessment.
The other thing I’m really excited about is the potential to reduce and remove barriers to participation in clinical trials. By taking the technology to the patient, instead of bringing the patient to the site, we can hopefully increase the diversity of people that are included in clinical trials. That should improve the generalisability of the results that we get from clinical trials.
I’m really excited about the potential to reduce and remove barriers to participation in clinical trials. By taking the technology to the patient, instead of bringing the patient to the site, we can hopefully increase the diversity of people that are included in clinical trials.
Your talk in the session is about automating voice-based cognitive testing. What are the advantages of this approach?
This is something that we’ve been working on at Cambridge Cognition for quite a few years now. We have worked to develop something that’s very robust, usable, and useful for clinical trials. The way that we’ve approached this has been by taking active assessments based on very well-understood cognitive paradigms and trying to replicate the level of automation that you would get with one of our CANTABTM tasks. We have invested a lot of work into getting the stimuli and the user interaction of these tasks just right, using a range of approaches including natural language processing, and automatic speech recognition tailored to cognitive testing. Our aim is to make the administration and scoring fully automated, and to capture voice data for later analysis. This provides us potentially additional meaningful signals, capturing how participants voices change over time or in response to task conditions.
Having robust automated cognitive assessments that can be deployed at scale and that don’t rely on people coming into clinics or needing specialised equipment, combined with machine learning voice biomarkers is a really exciting proposition. I think it could really change the way in which we can deliver verbal assessments in clinical trials.
What do you like to do outside of your working day?
Digging! I’m a keen gardener. I also chose the hottest day of the year to try and make two litres of jam, which maybe wasn’t the best idea!