Dr Jane M Rondina is currently a Research Associate at the Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology at University College London. She obtained a BSc in Computer Science and an MSc in Computer Engineering, applying image processing techniques to segment anatomical structures in cardiac MRI. She obtained her PhD from the Department of Neurology, Medical Sciences Faculty, Unicamp, Brazil, where she worked with functional MRI using both verbal and visual stimuli to study memory in patients with different etiologies of epilepsy. In her post-doc experiences in KCL and UCL, she has worked with machine learning (ML) techniques applied to neuroimaging and other types of clinical data. She also developed a new method (SCoRS) to select the most relevant features from brain imaging based on the theory of stability selection. Also, she has obtained experience with software development for brain imaging analysis, being part of the teams that developed toolboxes specifically for ML applied to neuroimaging (Probid - Pattern Recognition of Brain Image Data, in King's College London and PRoNTo – Pattern Recognition for Neuroimaging Toolbox in UCL). Through collaboration with clinical researchers from different groups, she acquired experience in the application of different ML methods involving data from different clinical conditions. She has also led investigations addressing different approaches to extract relevant features from images with lesions. Her current research interests include integration of information from multiple sources of data (involving different modalities of neuroimaging, clinical, demographic, and genetic data, as well as different ways of extracting features from each kind of data) in order to extract biomarkers to predict response to treatment of neurological and psychiatric diseases. In line with this objective, she is investigating unsupervised learning approaches to stratify patients in specific domains, ultimately aiming to enable the indication of therapies with greater potential for success based on individual multi-source data.