Machine Learning Data Scientist

Sara Elgayar

Clovis, CA 585-***-**** ****.*******@*****.*** linkedin.com/in/sara-elgayar-55685339 Profile

Computer Science and Data Science professional with experience in teaching, research, and applied AI. Began academic career as a Teaching Assistant at the British University in Egypt, mentoring diverse student populations in computer science. Completed an M.S. at the University of Rochester with a focus on Data Science and Machine Learning. Currently a Data Scientist at the VA Geriatrics and Extended Care Data Analysis Center (GECDAC), applying machine learning, natural language processing, and large-scale data analytics to healthcare data. Skilled in Python, R, SQL, and SAS, with expertise in algorithms, AI models, and multimodal data analysis. Demonstrated ability to bridge theory and practice through research and the development of data-driven tools that serve healthcare. Industry & Research Experience

Data Scientist / ML & AI Scientist — VA Geriatrics & Extended Care Data Analysis Center (GECDAC) Feb 2024 – Present

• Developed XGBoost classifier for 6M+ patients for Long-Term Institutionalization (LTI) prediction (AUC: 0.92), outperforming logistic regression by 15%.

• Built an LLM-powered NLP pipeline (OpenAI API, LLAMA3) to analyze clinical patient notes, automating caregiver identification for ADL needs (Accuracy 0.91 compared to 0.89 using traditional machine learning methods).

• Designed SQL-driven dashboards and ArcGIS visualizations for 26 million records of patients to optimizing resource allocation for senior leadership using PowerBI. Data Analyst / Biostatistician — University of Alabama at Birmingham, School of Medicine Aug 2023 – Jul 2025

• Analyzed data from the REGARDS cohort of 10,464 participants, examining the association of lower urinary tract symptoms and cognitive impairment using R studio, SAS, and Power BI, enhancing patient care efficiency

https://www.auajournals.org/doi/abs/10.1097/01.JU.0001008676.21744.5f.11.

• CARDIA-PLUS study: applied statistical modeling (Linear Mixed Models, survival analysis) to assess BMI impact on urinary health (published in Journal of Gerontology).

• The RISE for Health Study: Methods for In-Person Musculoskeletal Assessment https://www.ics.org/Abstracts/Publish/1128/000776_poster_20240924_105043.pdf

• The RISE FOR HEALTH study, a regionally-representative population-based prospective cohort study conducted by the Prevention for Lower Urinary Tract Symptoms (PLUS) Research Consortium, included a survey and in-person assessment designed to capture social, physical and biological data potentially associated with bladder health https://pfdweek24.eventscribe.net/fsPopup.asp?efp=SEJFVkpKU1kxOTE3MQ&PresentationID=1 450795&rnd=0.5248684&mode=presInfo

Page 2

Education

University of Rochester — M.Sc. in Data Science (GPA: 3.57/4.0) Aug 2021 – Dec 2022

• Coursework: Data Mining, Database Systems, Computer Vision, Statistics I & II, Time Series Forecasting, Tools for Data Science.

Ain Shams University — M.Sc. in Computer Science, Computer Vision & Machine Learning Jun 2009 – Jun 2013

• Thesis: “Vision-Based Topological SLAM for Autonomous Robots” Ain Shams University — B.Sc. in Computer Science, ML/AI department Jun 2009

Teaching & Academic Experience

Teaching Assistant — The British University in Egypt Sep 2009 – Sep 2012

• Taught introduction to programming in python, algorithms, and data structures to 60+ undergraduate students.

• Supervised lab sessions in image processing, data mining and artificial Intelligence.

• Assisted in developing instructional materials, assignments, and exams. Research Assistant — Rochester Institute of Technology Sep 2018 – Sep 2019

• Designed a Brain Imaging Data Structure (BIDS) pipeline for the ‘ExploreASL’ MRI toolbox.

• Integrated computational methods and algorithmic models into neuroscience research. Publications

A. Markland, A. D., Burgio, K. L., Williams, B., Kennedy, R., Elgayar, S., Zhang, Y., & Howard, V. J.

(2024). MP75-11 LOWER URINARY TRACT SYMPTOMS AND CORONARY HEART DISEASE RISK AMONG PARTICIPANTS OF THE REASONS FOR GEOGRAPHIC AND RACIAL DIFFERENCES IN STROKE (REGARDS) COHORT STUDY. Journal of Urology, 211(5S), e1236. https://doi.org/10.1097/01.JU.0001008676.21744.5f.11 (Original work published May 1, 2024) B. Meister, M. R., Elgayar, S. M., Fitzgerald, C. M., Fok, C. S., Lacoursiere, D. Y., Kenton, K., Low, L. K., Lukacz, E., McGwin, G., Norton, J., Newman, D. K., Smith, A. L., Wyman, J. F., & Sutcliffe, S. (2024, October 25). Prevalence of Pelvic Floor Myofascial Pain in a Cohort Study of Adult Women [Rapid Fire Oral Presentation]. PFD Week 2024.

C. Newman, D. K., Wyman, J., Sutcliffe, S., McGwin, G., Elgayar, S. M., Fitzgerald, C. M., Kenton, K., Lowder, J., Lacoursiere, D. Y., Low, L. K., Lukacz, E. S., Meister, M. R., Rodriguez-Ponciano, D. P., Smith, A., & Vaughan, C. P. (2024, September). Pelvic Floor Muscle Function in a Population-Based Study of US Women [Poster presentation]. International Continence Society Annual Meeting. Abstract # 776.

D. Lower Urinary Tract Symptoms and Cognitive Impairment among Participants of the REasons for Geographic and Racial Differences in Stroke (REGARDS) Cohort Study B. Williams, 1 S. Elgayar,1 Y. Zhang,3 K. L. Burgio,3 R. Kennedy,3 V. J. Howard,3 M. Crowe,3 A. D. Markland.1,2 1. Geriatric Page 3

Medicine, The University of Alabama at Birmingham, Birmingham, AL; 2. Birmingham/Atlanta GRECC, Birmingham, AL; 3. The University of Alabama at Birmingham, Birmingham, AL. E. Elgayar, S. M. O. (2013). Developing of Simultaneous Localization and Mapping Algorithms for Robots (Master’s thesis, Ain Shams University, Faculty of Computers and Information Sciences, Computer Science Department).

F. Elgayar, S., Salem, M. A.-Megeed, & Roushdy, M. I. (2013). Two-Level Topological Mapping and Localization Based on SIFT and the Wavelet Transform. In Recent Advances in Telecommunications and Circuits (Proceedings of the 2nd International Conference on Circuits, Systems, Communications, Computers and Applications, Dubrovnik, Croatia, June 25-27 2013). G. Elgayar, S., Salem, M. A.-M., & Roushdy, M. I. (2013). Nested Graph Representation for Visual SLAM based on Local and Global Feature Processing. International Journal of Computer Applications, 78(16), 1–8. https://doi.org/10.5120/13604-1350

H. Clement, P., Castellaro, M., Okell, T. W., Thomas, D. L., Vandemaele, P., Elgayar, S., Oliver-Taylor, A., Kirk, T., Woods, J. G., Vos, S. B., Kuijer, J. P. A., Achten, E., van Osch, M. J. P., Detre, J. A., Lu, H., Alsop, D. C., Chappell, M. A., Hernandez-Garcia, L., Petr, J., & Mutsaerts, H. J. M. M. (2022). ASL-BIDS, the brain imaging data structure extension for arterial spin labeling. Scientific Data, 9(1), Article 543. https://doi.org/10.1038/s41597-022-01615-9

I. Invited Speaker, VA Geriatrics and Extended Care Data Analysis Center (GECDAC) Annual Retreat 2025 — “Introduction to Applied Machine Learning for healthcare data analysis.”

Contact this candidate