Developed a "Personalized prediction model for hypoglycemia events" in diabetic patients, integrating machine learning algorithms with patient-specific data to enhance healthcare outcomes.

• Compiled and preprocessed diverse datasets, including continuous glucose monitoring (CGM) data, insulin dosages, dietary intake, physical activity levels, and patient demographics, ensuring comprehensive input for predictive modeling.

• Engineered advanced features to capture temporal patterns, glycemic variability, insulin sensitivity, and other relevant factors, improving the model's predictive accuracy and robustness.

• Implemented machine learning algorithms such as Random Forests, Gradient Boosting, and Deep Neural Networks to predict hypoglycemia events with high sensitivity and specificity.

• Applied advanced techniques, including cross-validation, hyperparameter tuning, and ensemble learning, to optimize model performance and ensure generalization across diverse patient datasets.

• Validated the predictive model using independent datasets, employing rigorous evaluation metrics such as receiver operating characteristic (ROC) analysis and calibration curves to assess model reliability.

• Collaborated closely with healthcare professionals to integrate the predictive model into clinical practice, enabling early intervention and personalized management strategies for patients at risk of hypoglycemia.

• Contributed to advancing the field of diabetes care by developing a robust and interpretable predictive tool, empowering clinicians and patients with actionable insights for proactive hypoglycemia prevention.

Digital Skills: Python, Machine Learning, Data Analytics, Model Development, Healthcare Informatics, Software Development.