Virginia Fernandez

Can you tell us about your PhD?

My PhD is trying to address the data scarcity that we have in medical images, especially the lack of some annotated data. We find there is data missing from images such as MRI brain scans, because it takes time to process and annotate them.

If we can create this data synthetically, then we might be able to better train AI to support clinical diagnoses. In the longer term, this could help support clinicians to make a diagnosis, and save them time.

Why did you choose to do a PhD at King's?

It is a funny story. I was finishing my first master's without any clue of what I wanted to do next. My master's supervisor was a board member of the interview panel at King’s, and he asked if I wanted to do a PhD. My answer was no because everyone I saw doing a PhD was miserable.

He suggested I apply anyway, and I got an interview. It was a full day interview, with a small journal club, a chance to present your work and an interview panel. We were also shown around and introduced to people. This completely changed my mind, because I loved the vibe.

They were also interested in my cartoons and drawing. They told me they really value public engagement and people with this profile.

King’s also felt collaborative and there were opportunities to work with clinicians and industry. Then when I was at my supervisor choice meeting, me and my supervisor clicked straight away.

It felt like the way he worked and the way he saw things was very complementary to the way I worked and saw things. By then I was convinced, and here I am.

Your PhD included a master's course in the first year, how was that?

I had already done a master's at another London university, so initially I was quite reluctant to do another. But the content was very useful and it helped me a lot. It was also a very different experience to my first master's where there were 200 students on the course.

The King’s master's course was more intimate, there were fewer people in the classes and the teachers were more motivated. It meant I was in a better place starting my PhD and when it came to writing my thesis.

Have you had opportunities to do any public engagement work during your PhD?

King’s has an on-site public engagement team, and they send a lot of opportunities around. So I met them when I first started and explained I like doing cartoons about my research. Quite quickly they asked me to do big panels explaining AI which they shared on the King’s website and has been used in publications.

I have also been drawing the views of patient groups from other people’s research. I try to get everyone’s feedback, opinions and comments into the drawing to summaries the patient group outputs. I have also had the opportunity to represent King’s Doctoral College at the Great Exhibition Road Festival in South Kensington, where we had lively conversations with the public about AI and health.

How do these interests sit with your PhD?

I think they are a great mental health break from working on a research topic. You also get to get your research out there and see people engage with it, and that is really valuable. It helps you understand where your research fits in society and gives you even more motivation for your topic.

With AI there is so much misinformation, and people can be afraid of it. I hope my work helps explain AI is a tool, and that it is nothing to be afraid of. It is so helpful that my supervisor understands and supports my work in this area.

Can you tell us about any networking and social events that are open to postgraduate students?

We have the Doctoral College at King’s that has lots of different groups and runs seminars, networking and social events. 

These feel even more important because I started my PhD in 2020 when everyone had to stay at home.

The networks I, and others in my Department, have developed has resulted in some interesting research collaborations in the last years of my PhD. These wouldn’t have happened if we hadn’t got to know these people. We built an open source repository to help other researchers with code that could be shared and reused. It is called MONAI Generative Models and has had interest from the big GPU company NVIDIA. We couldn’t have developed this without collaboration with different people in and outside my department.

Have you done any training courses during your PhD?

I attended a lot of courses at the beginning of my PhD as it was four years, with an MRes at the beginning. One of the most useful was on advanced machine learning, which was a deep dive into machine learning methods. Without that, I am positive that it would have taken me months before I got going with this for my PhD.

Another course I found useful was critical skills, which was a journal club. The person who ran it made it very dynamic with a host and an interesting discussion. It helped me to read papers quickly and find the most important information in them, including understanding the study strengths and limitations.

Is there anything that has surprised you about doing a PhD?

I didn’t think it would be this enjoyable. I think this is because I had great supervision throughout, great collaboration and a lot of opportunities, and I have done so much more than my thesis.

I have been able to learn things valuable to my career, and I have been happy. That is not to say I have not had imposter syndrome from time to time, but these have been happy years.

What advice would you give someone who was choosing a PhD now?

I have seen people who are only thinking about the research work when they are applying – their priorities are the best university possible, the best title or this exact topic. I think it is just as important to find a good supervisor and a good team that you feel comfortable with. My supervisor asks potential postgraduate students to meet not only him but his team. Because he understands it is the team that is important during a PhD.

What advice would you give to someone doing a PhD?

When you start it is easy to think you need to work 13 hours a day to demonstrate you are working hard – that is what everyone thinks a PhD is. Then you see people getting burnout after three or four months. I had that, and it is 100% not worth it, because it is not going to make your work go faster. It is important to slow down, and also get out there and go to events and do different things outside your research. They help you handle everything better.

What are you going to do after your PhD?

I am hoping to do a Post-Doc with my team, I need to have an interview but I am hopeful I will get the position. The new project will be on a different topic more applied to clinical use and will complement my skills. I like the idea of seeing what it is like to deploy an AI algorithm into healthcare, and deal with clinicians and project partners. Last summer I did an industry internship, so maybe in the future I would like to work in industry as well.

What has been your proudest PhD moment?

When my first paper was published. I was at a workshop in a conference and it got an award for the best paper. I published quite late compared to other people, and it was at the peak of my unhappiness as a PhD student. I wasn’t feeling very confident, I was not sure where I was going and imposter syndrome was creeping in. So getting a piece of research out there and having people value it, that was really good.

Whenever he saw I was struggling, he just grabbed me by the shoulders and said ‘look at what you're doing. This is good.’ I think a good supervisor spots these things in students and helps them.