About Me
Yeganeh Abdollahinejad
Machine Learning & Computational Science Researcher · Michigan State University
I'm a researcher working at the intersection of machine learning, computational modeling, and data science. I'm drawn to problems where the data is messy, the stakes are real, and the model has to earn trust — health, human signals, and information integrity.
My recent work spans multimodal misinformation detection — building models that reason across text and images and stay robust as stories evolve — and affective computing, where I design EEG and physiological-signal models for emotion and fatigue recognition that generalize to people they've never seen. Alongside that, I build mechanistic and Bayesian models for oncology and nanoparticle transport, and explainable pipelines for medical imaging and clinical data.
A thread that runs through all of it: I care less about chasing leaderboard numbers and more about whether a model is calibrated, interpretable, and honest about what it doesn't know — especially when the answer could affect a patient or a public conversation.
This site is where I keep my research, projects, writing, and the things I'm currently obsessed with. Take a look at what I'm chasing or browse my projects.
Research interests
Trustworthy & multimodal ML
Misinformation detection, cross-modal reasoning, and models that stay calibrated and honest under real-world shift.
Human signals & affective computing
EEG and physiological signals for emotion and fatigue recognition that generalize to unseen people.
Computational modeling for health
Mechanistic models, Bayesian inference, and ML for oncology, nanoparticle transport, and clinical decision support.
Explainable & responsible AI
Saliency, uncertainty, and fairness — making model decisions legible, especially in medicine.