PeopleofSuperintelligence

Researcher

Neel Dey

Neel Dey is a biomedical machine-learning researcher who develops computer-vision and representation-learning methods for medical imaging. He earned a master of science degree in electrical engineering at NYU in 2017, defended a doctoral thesis on medical image analysis in 2022, and received NYU's Pearl Brownstein Doctoral Research Award. His 2020 group-equivariant GAN preprint introduced symmetry priors to improve conditional image generation, and a brain MRI motion-correction paper he coauthored won MIDL's Best Oral Presentation award in 2023. In 2025, he coauthored Anatomix, which learns general-purpose biomedical volume representations from synthetic data, and in September 2026 he co-senior authored a Nature paper on xvr for surgical image registration. As of October 2026, he is an investigator and faculty member at the Martinos Center for Biomedical Imaging at Harvard Medical School and Massachusetts General Hospital.

Works
55
Citations
512
h-index
12

Publication numbers from OpenAlex, as of 27 October 2026.

Timeline

  1. 2017

    An M.S. at NYU

    He earns an M.S. in Electrical Engineering at New York University in 2017. His graduate research focuses on geometric machine learning and computer vision for medical and biological images.

  2. 2020

    Group equivariant GANs

    He posts Group Equivariant Generative Adversarial Networks as a 2020 preprint. The paper incorporates symmetry priors into GAN architectures through group-equivariant convolutions and improves conditional image generation in limited-data settings.

  3. 2022

    PhD research and a CSE award

    He receives the CSE Pearl Brownstein Doctoral Research Award in April 2022. In June 2022, he defends his Ph.D. thesis on medical image analysis and deep learning at NYU.

  4. 2023

    MRI motion correction at MIDL

    He coauthors Data Consistent Deep Rigid MRI Motion Correction, a deep-learning method that reconstructs motion-free images from motion-corrupted MRI. In 2023, the paper wins MIDL's Best Oral Presentation award.

  5. 2025

    Anatomix for biomedical volumes

    In 2025, he coauthors Learning General-Purpose Biomedical Volume Representations using Randomized Synthesis for ICLR. The anatomix model learns general-purpose features for 3D biomedical images from synthetic data.

  6. By 2026

    Postdoctoral work at MIT

    His postdoctoral work at MIT CSAIL's Medical Vision Group with Polina Golland centers on representation-learning frameworks for generalization across new tasks, biomedical contexts, and datasets.

  7. September 2026

    X-ray volume registration

    In September 2026, he co-senior authors a Nature paper on xvr, an AI system that matches a patient's X-rays with preoperative 3D scans. It adapts to each patient in about five minutes and performs the registration in seconds.

  8. As of October 2026

    Investigator at Martinos Center

    As of October 2026, he is an investigator and faculty member at the A. A. Martinos Center for Biomedical Imaging at Harvard Medical School and Massachusetts General Hospital. He builds machine-vision methods that biomedical practitioners can use as-is to understand their data.

Career

  1. NowMass General BrighamInvestigator and Faculty Member, A.A. Martinos Center for Biomedical Imaging
  2. NowHarvard UniversityInvestigator and Faculty Member, A. A. Martinos Center for Biomedical Imaging
  3. MIT CSAILPostdoctoral Researcher

Education

  1. 2017New York UniversityM.S. in Electrical Engineering
  2. New York UniversityPh.D.

From the biography.

Checked career sources: vida.engineering.nyu.edu, neeldey.com.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Neel Dey yet. Send us a free-licensed photo.

  1. Neel Dey | NYU Tandon School of Engineering engineering.nyu.edu. Checked 27 October 2026.
  2. Group Equivariant Generative Adversarial Networks arxiv.org. Checked 27 October 2026.
  3. VIDA – Visualization Imaging and Data Analysis Center vida.engineering.nyu.edu. Checked 27 October 2026.
  4. MIT researchers combine deep learning and physics to fix motion-corrupted MRI scans news.mit.edu. Checked 27 October 2026.
  5. anatomix: Stable Representations For Any Biomedical Image neeldey.com. Checked 27 October 2026.
  6. Neel Dey neeldey.com. Checked 27 October 2026.
  7. New AI technique could make minimally invasive surgeries safer and more precise news.mit.edu. Checked 27 October 2026.