Leveraging cutting-edge deep learning architectures, I developed an intelligent system for early detection of Diabetic Retinopathy using
retinal fundus images. Through advanced preprocessing techniques and the application of CNN models such as ResNet50, EfficientNet and Vision
Transformers (ViT), I built a highly accurate, interpretable diagnostic tool. The final solution was deployed as a real-time web application
using Hugging Face, showcasing not only the power of AI in ophthalmology but also my ability to design accessible, life-impacting healthcare
solutions.
Employing state-of-the-art techniques, I leveraged Support Vector Machines (SVM) and K-Nearest Neighbors (KNN)
algorithms to predict an individual's risk of a heart attack. By meticulously analyzing lifestyle and health factors of an individual, I crafted a robust predictive model that not only showcases the power of machine learning but also underscores my expertise in utilizing advanced algorithms for real-world, life-saving applications. This project not only demonstrates my proficiency in data science but also my commitment to addressing critical healthcare challenges through innovative and impactful solutions.
In a meticulous data refinement endeavor, I employed Power BI to perform comprehensive data cleaning using the expressive M language. The project focused on transforming a raw Superstore sales dataset into a streamlined and efficient form. Leveraging Power BI's capabilities, I crafted a robust Star Schema model, optimizing the dataset for enhanced analytical capabilities. Through the strategic application of Data Analysis Expressions (DAX), I generated dynamic visualizations that provided meaningful insights into the Superstore sales trends. This proficiency in data preparation, modeling, and visualization exemplifies my commitment to delivering actionable intelligence through advanced analytics.