Vision mamba for classification of breast ultrasound images
A. Nasiri-Sarvi, M. S. Hosseini, and H. Rivaz, “Vision mamba for classification of breast ultrasound images,” arXiv preprint arXiv:2407.03552, 2024.
CV
Education
- M.Sc. in Computer Science, Concordia University, 2023-2025 (expected)
- B.Sc. in Computer Engineering, Ferdowsi University of Mashhad, 2022
- Thesis title: “Thesis Title: Lightweight FAQ Answering Bot Using Siamese Networks”
Work experience
- Fall 2023 – Present: Graduate Research Assistant
- Concordia University
- Duties include: Developing efficient deep learning solutions for high-resolution images, focusing on both theoretical and system-level improvements; self-supervised algorithms for Gigapixel Whole Slide Images (WSI) addressing complex tissue structures and label scarcity
- Spring 2022 – Summer 2023: Machine Learning Researcher
- Motometrix Inc
- Duties included: Developed a physics-based simulation for LiDAR sensors, with a primary focus on modeling ‘raydrop’; designed a pipeline for processing LiDAR data to generate 3D vehicle assets; created a pipeline for generating 3D models of cities using satellite imagery and OpenStreetMap data; proposed a novel pipeline to detect occlusions for LiDAR sensors
- Fall 2020 – Winter 2022: Machine Learning Engineer
- ParsTech AI
- Duties included: Developed virtual try-on GAN models for high-resolution images using knowledge distillation; worked on disentangling and manipulating the latent space of StyleGAN2 to create a facial attribute editor application
- Fall 2020: Machine Learning Intern
- ParsTech AI
- Duties included: Developed deepfake image detection models using CNNs, transfer learning, and Fourier domain analysis
Publications
Vim4Path: Self-Supervised Vision Mamba for Histopathology Images
A. Nasiri-Sarvi, V. Q.-H. Trinh, H. Rivaz, and M. S. Hosseini, “Vim4path: Self-supervised vision mamba for histopathology images,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops
Presentations
Vim4Path: Self-Supervised State Space Modeling for Histopathology Images
Oral Presentation at IEEE/CFV Conference on Computer Vision and Pattern Recognition - CVMI Workshop, Seattle, WA
Teaching
Skills
- Programming
- Highly skilled in Python
- Familiar with Java, C/C++, Triton/CUDA
- Frameworks/Libraries
- PyTorch
- OpenCV
- NumPy
- Pandas
- Matplotlib
- scikit-learn
- Open3D
- ROS
- Embree
- pytest
- FastAPI
- Selenium
- Miscellaneous
- Linux (Ubuntu)
- Shell (Bash)
- LaTeX
- Gazebo
- Git
- Jupyter
- Docker
- VS Code
- AWS (EC2)