About
Physician and imaging AI researcher. I build the tools my research needs.
I am a physician and researcher at Tehran University of Medical Sciences. Most of my work circles around one question: how much can MRI actually tell us about inflammatory bowel disease. Not just whether disease is present, but whether it is active or chronic, reversible or fixed. Answering that has pulled me into quantitative imaging, applied AI, and eventually into building the tools myself, because the off-the-shelf ones were not quite right.
The longer version of how I got here is here.
Path
Jan 2024 — Present Research Associate, Abdominal Imaging and AI Department of Radiology, Shariati Hospital, Tehran University of Medical Sciences. Supervisor: Prof. Amir Reza Radmard.
The core work is the UC cohort: MRE features, bowel wall activity, motility from GIQuant, chronicity markers, validated segment by segment against endoscopy. Getting the modelling right meant moving from logistic regression through several ML approaches to Bayesian multilevel models, which handle clustered patients, ordered outcomes, and individual-level uncertainty in a way simpler methods could not.
The ideas kept expanding. Reports held structured data no one was extracting, so I built LLM pipelines to do that. Motility and wall features belonged with labs and clinical notes, so I started on multimodal models. Each step came from a gap the previous step revealed.
Ongoing Motilent · University College London, UK Quantitative colonic motility MRI in ulcerative colitis, in collaboration with Dr. Alex Menys and Dr. Iyad Naim (Motilent, London). The work applies GIQuant-derived displacement metrics to a UC cohort validated segment by segment against Mayo Endoscopic Subscores, with a focus on the relationship between motility, disease activity, and chronic structural damage. Two oral presentations accepted at ESGAR 2026. Manuscript in preparation for submission.
Ongoing Icahn School of Medicine at Mount Sinai, New York Prof. Jean-Frédéric Colombel (Icahn School of Medicine at Mount Sinai) is a co-author on the motility MRI manuscript, contributing clinical expertise and manuscript review.
Aug 2022 — Aug 2024 General Practitioner · Valiasr Local Medical Center Part-time clinical work in Alborz during the transition into full-time research. This is also when I started teaching myself Python properly: DataCamp, then 3Blue1Brown, then building things until they worked.
Sep 2022 — Dec 2023 Research Assistant, Imaging and Clinical Data Science Shahid Beheshti University of Medical Sciences. CT and MRI projects: protocol review, dataset construction, statistical analysis. Contributed to PanCanAID-related pipelines including curated datasets and semi-automated CT segmentation workflows.
May 2020 — May 2022 Medical Internship Teaching hospitals of Shahid Beheshti University of Medical Sciences, Tehran. Two years of rotations across internal medicine, surgery, paediatrics, obstetrics, emergency, and outpatient clinics. I thought I wanted to be a surgeon. I was wrong, but the clinical instincts stayed.
2015 — 2022 Doctor of Medicine (MD) Shahid Beheshti University of Medical Sciences, Tehran. The girl who actively avoided research in medical school. People change.
Get in touch
The fastest way to reach me is by email. If you are working on imaging in IBD, on multimodal medical models, or on any of the methods above, I would like to hear from you.