EMSBot is a comprehensive machine learning tool designed to assist emergency medical providers in rapid and accurate diagnosis of commonly encountered conditions. The system integrates various models to address multiple medical diagnostic tasks, including image classification, audio classification, time-series prediction, and natural language processing.
-Burn Injury Image Classifier
-Chest X-Ray Image Classifier
-Heart Failure Prediction
-Diabetes Prediction
-Stroke Prediction
-Lung Auscultation Diagnosis
-Heart Auscultation Diagnosis
-Electrocardiogram Classification
-Wound Image Segmentation
-Medical Report Text Processor
-Neural Networks
-Logistic Regression
-Random Forest Ensembles
-Gradient Boosting
-Decision Trees
Languages: Python
Libraries: TensorFlow, Keras, scikit-learn, NumPy, pandas, Matplotlib, Seaborn, librosa, imbalanced-learn
Frameworks: TensorFlowJS, ReactJS, ExpressJS
Model Improvement: Currently working on improving the following models: heart auscultation classifier, EKG classification, wound image segmentation.
Web Interface: Deployment using TensorFlowJS, ExpressJS, ReactJS, allowing real-time interaction and integration into medical workflows.