Marie Zelenina

I am currently a postdoc at Dr Lauren Atlas' Affective Neuroscience and Pain Lab at NIH/NCCIH, where I apply data science and machine learning tools to videos of facial expressions people experiencing pain. The project is in collaboration with NIMH Machine Learning Core.

I defended my PhD with Distinction in June 2025!

CONTACT
[first].[last]@nih.gov
[first].[last]@gmail.com
LinkedIn
Twitter
Google Scholar
Github

Hi, I am Marie! I combine my expertise in machine learning (MSc in AI from the University of Edinburgh, UK) and neuroscience (MSc in Neurobiology from the University of Porto, Portugal), to approach meaningful challenges in the domain of health and medicine - particularly in mental health and pain research. Depression is one of the main causes of disability worldwide (WHO). Pain affects more people than diabetes, heart disease, and cancer combined. (CDC). Depression and pain often occur together, and amplify each other. Treating comorbid pain and depression is a particular challenge.

My depression research has focused on the measurement of psychopathology, particularly the relationship between psychiatric assessments, underlying psychological constructs, and patients' experiences. During my PhD at NIMH, I combined psychometric methods and latent variable modeling, including work on hierarchical models of psychopathology such as HiTOP. I worked with multiple large datasets to validate depression measurements and gather insights in topics such as informant discrepancy (parent-child disagreement on whether the child is depressed).

My other work inlcuded multi-echo MRI data collection and analysis, with the goal of improving the targeting of Transcranial Magnetic Stimulation (TMS) for treatment-resistant depression (pilot dataset), and neuropsychoendocrinology: the effect of oxytocin on human EEG and its role in various psychiatric disorders.

I have prior experience in Computer Science and Artificial Intelligence/Machine Learning. For my thesis work for my MSc in AI (University of Edinburgh), I worked with eyetracking data and machine learning for a Natural Language Processing task. Before that, I got an undergraduade degree in a Computer Science-related field, where I applied neural networks to static images.