Aliasghar Mortazi, Ph.D.
Ali is a computer vision scientist with a passion for developing optimized A.I. methods for medical image analysis and deep learning explainability. Prior to joining Volastra, Ali worked as a senior A.I. engineer at Clario, developing efficient deep learning solutions for analyzing large scale medical images including CT and MRI. His methods were used to automate quality control, anatomical recognition, and sequence detection from multi-modal images in real world applications. He completed a postdoctoral fellowship at the Radiology department of University of Pennsylvania where he developed an automated robust method for disease quantification from FDG-PET/CT images of patients with skin disease such as cutaneous lymphoma or psoriasis, and thoracic cancer, for pretreatment planning and treatment response assessment. In addition, he developed a standardization method of FDG-PET images which mitigated the undesired variabilities in tissue radiotracer activity. The metabolic activity measured from FDG-PET is an important biomarker that is clinically used for diagnosis, staging, prognostication, and treatment response assessment purposes in patients with cancer. Ali completed an internship at Harvard Medical School and Boston Children’s Hospital working on automating DCE-MR reconstruction methods by segmenting aorta automatically and accurately. His method was used in the pipeline for estimation of glomerular filtration rate, the best indicator of kidney function for patients with compromised renal function. He also developed deep learning based methods to detect and segment Crohn’s disease from pediatric MR images. He received his Ph.D. in computer science from University of Central Florida’s Center for Research in Computer Vision, where he focusing on developing A.I. and deep learning methods for medical image analysis. In his Ph.D. thesis, he introduced state-of-the-art optimum deep learning systems for medical image segmentation. He received his master’ in Electrical Engineering from Sharif University of Technology, where he developed artificial neural networks and reinforcement learning methods for robot training.