Maad Ebrahim

- Research, design, and implement robust classification, segmentation, and landmark (points and axes) detection methods for hip X-ray fluoroscopy using synthetic digitally reconstructed radiographs (DRRs) from CT scans.
- Mitigate domain shift between synthetic DRRs and real intraoperative X-ray images to enable reliable deployment in operating rooms. This was achieved through systematic evaluation of neural network architectures, normalization strategies, style transfer, and domain adaptation, along with careful optimization of data augmentation types and parameter ranges.
- Leading the development and integration of the machine learning pipeline into the complete product pipeline. The pipeline aims to automate intraoperative assessments using classification, segmentation, and landmark detection of one or more X-ray images to later perform 3D reconstruction, registration, and pose correction to provide the required surgical measurements for Total Hip Surgery (THA).

• Collaborated with interdisciplinary teams of data scientists to design, implement, and patent an intelligent Reinforcement Learning (RL) based Coordinated Multi-Point (CoMP) scheduler for 5G Cloud Radio Access Network (RAN) networks.
• Improved the Deep RL agent training performance in mission-critical networks using a Quantum-Inspired Machine Learning-based optimizer.

• Towards Artificial General Intelligence; Scaling, Emergence, and Alignment (IFT 6760A):
Mentored graduate students on research projects and evaluated final reports.

- Proposed an industry-academia collaborative project between the company and the University of Montreal to build a multi-factor Authentication system for human-coupled IoT devices using state-of-the-art technologies, including machine learning, Blockchain, and built-in SIM security.
- Worked with my supervisor to write a grant proposal to the Natural Sciences and Engineering Research Council of Canada (NSERC), which was accepted for a 4-year funding to cover a couple of postdoctoral fellows and several graduate students.