Having not studied maths or sciences at A-level I was an unusual candidate for a STEM degree. I already had a background in care work and nursing, and before that, in music and arts administration. I had learnt early on in my first degree (music) that to survive in today’s job market the ‘job for life’ no longer existed. In fact, being a full-time musician was almost out of the question, and that even the best musicians teach for at least 40% of their working year. Paid work was very tough to find on graduation, being shortly after the 2008 financial crisis. This led me down a circuitous path where I decided to let go of my dream of being a professional violinist, and opened the door to many opportunities, some a better fit than others.
I had already caught the science bug long before in 2010, when I first started listening to the Infinite Monkey Cage with Brian Cox and Robin Ince on Radio 4. I listened to every podcast multiple times, and often hung around the magazine aisle reading a copy of New Scientist that had just been released. When I specialized in neurosurgical nursing, I applied for a clinical fellowship in the hospital’s Computational Oncology Lab (1) researching high grade glioma, a particularly lethal type of brain tumour. The project aimed to test machine learning algorithms on clinical MRIs of patients suffering from these brain tumours. Critically the imaging shown to the algorithm would be longitudinal, meaning it would show patients’ tumours over a period of treatment. The algorithm would learn from the human drawn segmentations of the tumour, then it’s learning would be tested on an unsegmented set of images. The newly trained algorithm could (hopefully!) segment the tumour automatically without any human interaction.
Automated segmentation using clinical prognosticators could save the time of many radiologists, speed up clinical diagnosis and thus patients may even have the chance of earlier treatment time. The next step, once achieving automated segmentation for longitudinal imaging, would be to predict where the tumour might recur.
The project was thought up by my supervisor, and was clearly an excellent idea, as Heidelberg University published results of a very similar project 1 year into my fellowship. The Heidelberg paper was fantastic and done on a much greater scale than my research could achieve, I recommend a read if you are interested (3). It also meant that I had to find a different niche within my existing project to ensure my research stood out. At this point COVID-19 stepped in and I went back to full time clinical work, while funding was redirected from research to clinical care.
Following the impact of COVID-19 I decided to pursue a Biomedical science degree, where I could learn the molecular basis underpinning my existing knowledge, and also learn statistical analysis so I could present my data professionally.
My friend suggested I go for the music scholarship at University of Kent which offered some funding towards each academic year. This meant auditioning, which I have always found absolutely terrifying. Luckily the music department staff Dan, Sophie and Flo were there to put me at ease and it is due to their dedication and creativity that music has flourished once again in the fantastic summer-music week of 2020. After 12 months of no concerts it felt unbelievable to stand there in front of an audience and get to play together. I will be forever grateful to the music department for that opportunity.
My biggest challenge came from the course itself which, for me, was the first time I had encountered chemistry and maths for many years. I felt I had to put my all into it otherwise I had no chance, and I managed to surprise myself by doing well in coursework assignments. Thanks to my tutors I was starting to feel like a scientist, to think through a problem, to ask why and to use my developing knowledge as a stepping stone towards answering difficult questions.
In terms of advice, I am inclined to repeat what my supervisor told me, which is to become comfortable in living with uncertainty. He said that if you want to be in research, uncertainty is par for the course, but very early in school we encounter scientific ‘certainties’ on which we develop a huge base of our knowledge, thanks to the work of pioneering researchers before us. This leads us to expect certainty from our work, and can lead to frustration when it is not found.He said to not give up hope because when we embrace uncertainty, our mind stays open to all possibilities.This leads to more creativity in our thought process and perhaps,if we are very lucky, a discovery.
Kamnitsas K, Ledig C, Newcombe VF, Simpson JP, Kane AD, Menon DK, Rueckert D, Glocker B. Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Medical image analysis. 2017 Feb 1;36:61-78. Available online at: https://www.sciencedirect.com/science/article/pii/S1361841516301839
Kickingereder P, Isensee F, Tursunova I, Petersen J, Neuberger U, Bonekamp D, Brugnara G, Schell M, Kessler T, Foltyn M, Harting I. ‘Automated quantitative tumour response assessment of MRI in neuro-oncology with artificial neural networks: a multicentre, retrospective study’. The Lancet Oncology. 2019 May 1;20(5):728-40. Available online at: https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045%2819%2930098-1/fulltext
Daniel Harding, ‘In pictures Summer Music Week’. 2020 June. Available online at: https://www.kent.ac.uk/music/news/10677/in-pictures-summer-music-week-2021
Thanks Kammy for taking the time to share your experience with us! If you’d like to share your story, get in touch with us by emailing email@example.com.