CV

Education

• Ph.D. in Biomedical Informatics, Arizona State University, 08/2024 – 08/2028
  • Advisor: Prof. Qiyun Zhu
  • GPA: 4.00
• MS in Electrical Engineering, University of Washington, 09/2022 – 03/2024
  • Mentors: Prof. Stan Birchfield, Prof. Tamara Bonaci
  • GPA: 3.79
• BS in Electrical Engineering, University of Business and Technology, 09/2016 – 08/2021
  • Advisor: Prof. Mohammed Al-Qarni
  • GPA: 3.43

Honors and Scholarships

======

• SACM Scholarship (Saudi Arabia Cultural Mission) 2023-2024
• SACM Scholarship (Saudi Arabia Cultural Mission) 2022-2023

Research Experience

======

• Literature Review on Privacy-Preserving Machine Learning for EHR
  • University of Washington
  • Mentor: Professor Tamara Bonaci (January 2024 – March 2024)
  • Conducted an extensive literature review on privacy challenges in EHR systems, focusing on the intersection of machine learning and patient data privacy. Analyzed current strategies and identified gaps in the literature that could hinder the potential benefits of machine learning in healthcare privacy.
• Dataset Synthesis for Computer Vision with NVIDIA
  • NVIDIA
  • Mentors: Professor Stan Birchfield and Dr. Jonathan Tremblay (July 2023 – March 2024)
  • Developed a GitHub repository for synthesizing the YCB dataset using Blender, nvisii, and the YCB Dataset to support research on the LOFTR model. Gained practical experience in machine learning and computer graphics, focusing on data preparation and analysis for AI applications.
• Literature Review on Federated Learning in Adversarial Settings
  • University of Washington
  • Mentor: Professor Tamara Bonaci (June 2023 – March 2024)
  • Performed a comprehensive literature review on Privacy-Preserving ML techniques within the federated learning framework. Evaluated current methodologies and contributed to discussions on potential improvements and future directions in federated learning.

Projects

======

• Differential Privacy/Pseudonymization in Federated Learning for Medical Data
  • Implemented differential privacy and pseudonymization in TensorFlow Federated for protected medical data learning.
  • Crafted preprocessing pipeline for federated dataset creation and feature standardization/encoding.
  • Reduced MAE significantly while upholding data privacy, balancing model performance and patient information security.
• Development and Implementation of a Fact-Checking Model using NLP
  • Developed an NLP application for automated fact-checking as part of FactCheckNLP open-source project.
  • Trained claim classification models using PUBHEALTH dataset; achieved 80% accuracy with DistillBert.
  • Employed T5 and PEGASUS models for explanation generation, improving ROUGE scores for text summarization.
• GPT-3 API through Zero-Shot and Few-Shot Prompting in NLP
  • Led a NLP analysis using OpenAI’s GPT-3.
  • Applied zero-shot and few-shot prompting techniques on five SuperGLUE datasets.
  • Achieved an exceptional accuracy of 76.8% in the Relevant Strategy approach during few-shot learning.
• Deep Learning for ECG Arrhythmia Heartbeat Classification: A Comprehensive Guide + Kaggle Contest
  • Applied advanced deep learning techniques, achieving a maximum validation accuracy of 97.47% using a Convolutional Neural Network model.
  • Addressed class imbalance through random under-sampling and SMOTE, enhancing model stability.
  • Optimized logistic regression and random forest models, achieving accuracy scores of 76.5% and 92.2%, respectively.
• Stress Level Prediction using Heart Rate Variability and Machine Learning
  • Engineered and normalized features based on heart rate variability (HRV) for stress level prediction.
  • Evaluated multiple machine learning classifiers, including Decision Trees, Adaboost, and Random Forests, focusing on key performance metrics like F1-score and accuracy.
  • Conducted feature importance analysis using cross-validation, laying the groundwork for model improvements.

Teaching Experience

======

• Online Tutor, Varsity Tutors
  • PYTH 101: Python and Machine Learning (Fall 2020)
  • Taught nine students Python and Machine Learning three times a week, resulting in a 99% satisfaction rate and significant improvement in their understanding and application of the subjects.

Skills

======

• Languages & Abilities
  • Proficient in English and Arabic
  • Unrestricted Work Authorization in the United States and Saudi Arabia
• Programming Languages & Machine Learning Frameworks
  • Proficient in: Python (Sympy, LaTeX, NumPy, Pandas, Matplotlib, Seaborn, SciKit-Learn, Image, GANs, 2D and 3D GNNs, NLTK, Huggingface, OpenAI APIs, Tensorflow, Pytorch), Matlab, SQL
  • Familiar with: NVISSI, Blender, Blender Python API, C
• Tools & Packages
  • AWS Cloud Computing, Google Cloud Computing (VertexAI, AutoML), IBM Watson
  • Git, Anaconda, Jupyter Notebooks & Jupyterlabs, VSCode
  • Data Science Tools, Microsoft Office 365, Google Sheets, WordPress, Video Conferencing