I've always had a talent for fixing things, starting from when I was a child. My first sports car, which constantly broke down, gave me lots of chances to practice my DIY skills. I managed most repairs on my own, but when things got tricky, I knew when to call an expert. My friends like to joke, "Valery, you should have been a mechanic—forget statistics!" They even nicknamed me "Valery the DIY mechanic" because I could handle the simple fixes but wasn’t afraid to ask for help when needed. Even though I didn’t always get it right, this experience taught me the importance of persistence, learning as I go, and knowing when to seek help—a lesson I apply in every part of my life!

Growing up in the Southwest Region of Cameroon, I experienced firsthand the challenges of accessing specialised healthcare in underserved communities. This early exposure fuelled my drive to develop solutions that are both impactful and inclusive. My journey in multiple sclerosis (MS) research embodies this passion, with each project designed to deliver actionable insights into disease progression. These insights not only aim to advance treatment options but also to empower individuals with MS through tailored, patient-centered care pathways. One of my proudest moments was collaborating with an MS advocacy group to translate complex statistical findings into tools that could support informed decision-making. Knowing that my work could help ease someone’s journey with MS inspires me to continue pushing boundaries in research and innovation.

One of the most exciting developments in MS research is the use of advanced statistical models, such as multi-state Markov transition models, to estimate the effects of treatments and other factors on MS disability progression. These models enable us to study the dynamics behind each treatment and understand how a person's disability evolves over time. By integrating large real-world data and predictive modelling, especially within Bayesian frameworks, we are revolutionising MS care. This shift allows for tailored care, charting each individual's unique health journey with precision. My research aims to bridge clinical science and technology, enhancing personalised medicine in MS. This convergence of predictive analytics, innovative therapies, and patient-centered approaches marks significant progress in understanding and treating MS.

My current research is focused on pioneering innovative methods for personalised monitoring and management of disability in relapsing MS. By estimating individual frailties in disability progression through multi-state Markov modelling, I capture the complex and dynamic nature of MS, addressing both observed and unobserved heterogeneity among individuals. My work evaluates the effects of high- and low-efficacy disease-modifying therapies (DMTs) on transitions across MS disability states using continuous-time modelling. This approach moves beyond traditional binary outcomes, providing a more granular understanding of how therapies alter the risk of disability progression and potential improvement. Additionally, I incorporate biomarkers, imaging data, and real-world evidence to predict individual disability trajectories, ultimately aiming to create tools that enable personalised treatment plans. This research stands at the intersection of statistical innovation and clinical application, offering transformative potential for MS care and chronic disease modelling.

My research is critical because it tackles the complexity of MS progression, offering a more personalised approach to treatment. By using advanced statistical models, such as multi-state frailty models, I capture the unique, dynamic disability trajectories of people living with MS, providing deeper insights into how the disability evolves over time. This approach allows for a more precise evaluation of disease-modifying therapies (DMTs), particularly their effects on the risk of disability progression. By incorporating biomarkers, imaging, and real-world data, I aim to develop tools that enable clinicians to create personalised care plans for each person. This shift from traditional, generalised treatment approaches to individualised monitoring will help healthcare providers make more informed decisions, improving outcomes and reducing the burden of MS on individuals. Ultimately, my research has the potential to transform MS care by making it more tailored and responsive to each person’s unique disease trajectory.

I spend my time developing multi-state frailty models to analyse longitudinal disability data generated in MS studies. By incorporating frailties (random effects) in multi-state Markov transition models, my research accounts for unobserved, individual-specific characteristics that influence a person’s progression through disability states. These unmeasured factors might include genetic predispositions, immune system responses, or other personal traits that are not captured by standard clinical variables. This approach allows for a more nuanced understanding of how disability evolves in each person. After developing these models, I validate them and make them scalable to large datasets. Given the time-consuming nature of this work, I use high-performance computing (HPC) provided by our institution to optimise the models and estimate the frailties. Through this process, I have successfully estimated time-dynamic continuous frailty scores for each individual, which can identify individuals at risk for early gait impairment, enabling targeted interventions to mitigate disability progression.