Keerti Reddy Resapu

• With over four years of experience in healthcare data analysis and health informatics, I am passionate about leveraging data and technology to enhance patient outcomes and streamline healthcare operations. Pursuing a Master's in Health Informatics at George Mason University, I specialize in data collection, analysis, and visualization using Python, R, SQL, and Tableau,Weka.

• Data Mining
• Predictive Modeling

• Data Analytics Essentials
• Microsoft Excel
• Conflict of Interest
• Analyzing and Visualizing Data with Microsoft Power BI
• Python

Predictive Modeling on Excess Readmission Ratio (MIMIC Data)                                                              

• Integrated ICD codes filtered heart failure cases and calculated length of stay and admission intervals. Identified potential readmissions within a 30-day post-discharge window and merged data with patient demographics.
• Explored various machine learning models, including logistic regression, random forests, gradient boosting, and neural networks. Implemented k-fold cross-validation and oversampling techniques to ensure balanced training. Assessed performance using AUROC, accuracy, precision, and recall.

Predictive Modeling of Antidepressant Efficacy

• This project aims to predict the response to fluoxetine among African American participants in the All of Us database using python. By analyzing demographic and clinical factors, we seek to enhance personalized treatment strategies for this population. Analyze the demographics, treatment history, and clinical characteristics of African American participants prescribed fluoxetine.
• Fit a regression model to identify direct predictors of response to fluoxetine, utilizing the SAFE procedure to refine independent variables. Employ LASSO regression to predict fluoxetine response, reducing predictors to approximately ten. Key outputs will include regression coefficients and McFadden R-square values.

Data Visualization Project for Flowers Hospital

• Explored the impact of the Hospital Readmissions Reduction Program (HRRP) on Flowers Hospital in Alabama by analyzing Medicare payments and the associated readmission penalties. Utilized SQL, Microsoft Excel, and Tableau for data extraction, trend analysis, and visualization of excess readmission ratios across various conditions.
• Developed interactive dashboards to track hospital performance metrics and communicate findings to cross-functional teams for performance improvement initiatives.
• Implemented state-mandated reporting standards to ensure compliance with CMS regulations, focusing on excess readmissions for heart attack, heart failure, and pneumonia.

Fluoride Content Analysis and Fluorosis Risk Assessment

• Created a tool to assess fluorosis risk based on fluoride content in local water supplies, providing personalized recommendations for prevention.
• Extracted and analyzed water quality data using Python and Amazon Web Services (AWS), implementing algorithm development to predict fluorosis risk based on user-inputted postal codes.
• Developed a user-friendly interface and ensured project completion through project management and strategic planning methodologies.
• Presented findings through comprehensive data visualization in PowerPoint, demonstrating the relationship between fluoride levels and health outcomes.

AI-Driven Patient Monitoring: Assessing Depression Symptoms and Outcomes

• In this project, I developed a comprehensive system for tracking and evaluating depression symptoms in patients using the PHQ-9 questionnaire over a 4-month period.
• Key components included:
• Data Collection: Designed an online PHQ-9 assessment tool using qualtrics, ensuring HIPAA compliance and ease of access on both mobile and desktop devices.
• Monitoring System: Created an automated system to dispatch the PHQ-9 questionnaire at designated intervals, with alerts for unreceived or unanswered emails. Simulated responses to test the system’s effectiveness.
• Depression-Free Days Calculation: Analyzed PHQ-9 scores to calculate Depression-Free Days (DFDs), providing valuable insights into treatment efficacy and patient outcomes.
• This project underscores my expertise in healthcare analytics, data management, and patient monitoring, contributing to improved mental health management.