aa1823 / cardiostatsusa Goto Github PK

🔬 Who Am I?: I am a scientist that wears many hats. By training, I'm a Biostatistician with a Master's degree from California State University, but I also have a background in data analytics and data science. My journey in data analysis and statistical programming has been driven by a long-standing deep interest in public health epidemiology.

📊 My Experience: Previously, I've done modeling work to learn about the impact of diet on cardiovascular disease in US adults. My current focus: working with my lab to analyze the impact of certain xenobiotics on bronchopulmonary dysplasia in infants.

🌟 My Goal: My current goal is to improve and expand my statistical knowledge and skillset to better answer unsolved questions in the public health arena. I hope to one day give back and make great contributions to the field of biostatistics in a way that has a massive positive impact on people's health from all walks of life--whether locally or globally.

❤️‍🔥 My Hobbies: In my free time I like to learn new things, meditate, read, write, and analyze sports data. My current passion projects include: learning Python, studying African history, making art, and indoor-gardening.

Highlighted Projects

Paper | Repository

• Objective: To identify factors influencing chronic graft-versus-host disease onset in allogeneic bone marrow transplant recipients across 4 different clinical sites: The Ohio State University Hospitals, Hahnemann in Philadephia, St. Vincent's Hospital in Sydney, Australia, and Alfred Hospital in Melbourne.
• Analysis/Tools: Survival analysis techniques in SAS: Kaplan-Meier, AFT modeling.
• Outcome: Uncovered disparities in CGVHD onset across different demographics and clinical sites; St. Vincent patients experiencing a 35% longer time to CGVHD and Alfred patients experiencing a 10.3% shorter time.

Presentation | Repository

• Focus: Forecasted the full 82 game season performance of the Lakers using data from 73 games.
• Analysis/Tools: Applied linear discriminant analysis to predict win-loss records.
• Outcome: Forecasted that had the Lakers played all 82 games, they would have won 55 games, and lost 27, with a final win percentage of 67% and a loss percentage of 33%.

Presentation | Repository

• Focus: Developing a predictive model for the Lakers' points per game using 5 key features.
• Analysis: Multiple linear regression in R.
• Outcome: Model explained 86% of the variance in the team's performance.

Presentation | Repository

• Focus: Investigating the correlation between height and cardiovascular disease.
• Analysis: Examined a large dataset from Kaggle.
• Outcome: Identified a 47% probability of CVD in taller individuals.

Presentation | Repository

• Focus: Is there an observable difference in baseline creatinine levels and blood pressure between sexes.
• Analysis: Examined a kaggle dataset of electronic medical records of 491 patients from Tawam hospital in Al-Ain city in the United Arab Emirates in 2008 using SAS.
• Outcome: There is a difference in baseline creatinine across sexes, and it is statistically significant. On average, women over 20 µmol/L lower compared to men in the dataset. There was no effect of blood pressure on serum creatinine.