Mehrdad Salimitari

Bio:

I am an AI/ML research scientist and engineer working at the intersection of deep learning, computer vision, and medical imaging, with a background spanning medical AI, autonomous driving, energy systems, and production software engineering.

Currently I am a Postdoctoral Fellow at the University of Maryland School of Medicine, in the Trauma Radiology Artificial Intelligence Lab (TRAIL), advised by David Dreizin, MD. I built and deployed a patent-pending two-stage deep learning pipeline for traumatic hemorrhage detection in 3D CT that segments and classifies at both slice and patient level using mask-guided attention and a Vision Transformer, now in production at the R Adams Cowley Shock Trauma Center. I also developed a report-supervised 3D vision-language model that detects 83 distinct traumatic injuries from paired CT scans and radiology reports with no manual annotation, and a production LLM pipeline that structures 8,700+ free-text radiology reports into labeled training data, training at scale with PyTorch DDP on 8xH100 servers.

Previously, as a Postdoctoral Researcher at the University of Denver, I designed multimodal deep learning models for grid-induced wildfire risk prediction and virtual power plants, fusing satellite imagery, weather, and grid data.

Before that, I led backend engineering and cybersecurity for two consumer fintech platforms, managing a team of 8 within a 23-engineer organization and a 24/7 custodial wallet settling roughly $2M in daily transactions. I led the security remediation after a $1M breach and drove a 19-finding third-party audit to closure.

I received my Ph.D. in Computer Science from the University of Central Florida (UCF) under Prof. Yaser Fallah, focusing on computer vision and deep learning for autonomous driving and cooperative sensor perception. During my studies I also collaborated with the Center for Research in Computer Vision (CRCV) at UCF under Prof. Nazanin Rahnavard, developing sample/feature-selection algorithms for efficient deep learning.