Iman Aganj

Timeline

1. 2005–2011

   Computer science and MRI

   He earns a BSc in computer science from École Polytechnique in 2005 and an MSc and PhD in electrical engineering from the University of Minnesota, completing the doctorate in 2010. His doctoral work develops MRI methods for cortical-thickness measurement, diffusion-weighted fiber tracking, and high-resolution reconstruction. At the Martinos Center, he joins in 2011 and works on image registration and segmentation, diffusion MR analysis, and brain connectivity.

2. 2014

   An Alzheimer’s connectomics grant

   A recently promoted faculty member of the Laboratory for Computational Neuroimaging, he receives his first funding award in 2014 for a proposal on connectomic biomarkers of Alzheimer’s disease observed in multisynaptic pathways. The project is selected for funding through the Harvard NeuroDiscovery Center, Massachusetts ADRC, and MGH Neurological Clinical Research Institute’s 2014 Neurodegenerative Diseases Pilot Study Grants program.

3. 2021

   Expected label value segmentation

   In 2021, he and Bruce Fischl publish an expected-label-value method for soft medical-image segmentation. It estimates the expected label by considering possible atlas-to-image transformations, avoids deformable registration, and is evaluated on brain, liver, and pancreas MRI and CT datasets.

4. 2023

   Mapping Alzheimer’s connectomes

   In 2023, he and coauthors explore structural connectivity against clinical and demographic variables in four diffusion-MRI databases, including three related to Alzheimer’s disease. They identify 85 significant relationships using direct connections and 101 when they add multisynaptic pathways.

5. 2026

   Graph models for brain MRI

   In February 2026, he coauthors a graph neural network study of cortical-surface segmentation that combines structural MRI features with diffusion-MRI connectivity; the multimodal model reports a Dice score of 0.927. In May 2026, he coauthors a graph-based deep model for predicting demographic and clinical variables from structural brain connectivity.

6. As of October 2026

   Harvard radiology faculty

   As of October 2026, he is an assistant professor of radiology at Harvard Medical School and an investigator at the Martinos Center. His research covers medical-image registration and segmentation, diffusion-MR reconstruction and analysis, and brain connectivity; his 2026 work includes cardiac-MRI segmentation and graph-based connectome prediction.