HEALTHCARE PROJECTS
Pneumonia Classification using Chest X-Rays
Pioneered a sophisticated medical diagnostic solution using ResNet 50 architecture implemented in TensorFlow. Data augmentation techniques have been incorporated and the model uses Adam optimizer. Achieved a 91% accuracy in pneumonia detection through chest X-rays, showcasing a profound impact on advancing precision and efficiency in medical imaging diagnostics.
Detecting Parkinson's Disease using XGBoost
Designed and implemented an advanced predictive model to discern the presence of Parkinson's disease in patients, employing the UCI ML Parkinson's dataset as the foundational input. Employed rigorous hyperparameter tuning via GridSearchCV to meticulously optimize model configurations, ensuring optimal performance in disease diagnosis. Notably, the model achieved an accuracy of 94.87%, emphasizing its efficacy in aiding early detection and medical decision-making.
Predicting Heart Disease using Support Vector Machines
Leveraged the power of Support Vector Machines to construct a robust predictive model for the early detection of Heart Disease in patients. Employed meticulous hyper-parameter tuning through GridSearchCV to pinpoint and implement optimal model configurations, ensuring the highest accuracy in diagnoses. Furthermore, harnessed Principal Component Analysis (PCA) for effective dimensionality reduction and insightful visualization of the final results. Achieved an accuracy of 89.33%.