In medical emergencies, timely and accurate drug recommendations can significantly improve patient outcomes. The proposed Drug Recommendation System in Medical Emergencies using Machine Learning is designed to assist healthcare providers by suggesting optimal medications based on patient data and emergency scenarios. The system leverages machine learning algorithms to analyze patient medical history, symptoms, and vital statistics in real-time. The core of the system integrates advanced techniques such as Natural Language Processing (NLP) for processing unstructured medical records and classification models like Random Forest, Gradient Boosting, or Neural Networks for predicting suitable drugs. It incorporates a comprehensive drug database to cross-check contraindications, allergies, and potential drug interactions. The system is trained on a large dataset of medical records and emergency case studies, ensuring high accuracy and adaptability. Additionally, it provides a user-friendly interface for healthcare professionals, offering drug recommendations along with explanations to ensure transparency and trust. This solution aims to reduce decision-making time, minimize human errors, and enhance patient safety in critical situations, ultimately contributing to more efficient and effective emergency care.