This project presents the development of an intelligent real-time hypoglycemia prediction system designed to enhance diabetes management by providing accurate and timely predictions of hypoglycemic events. The system integrates deep learning algorithms within a mobile application that interacts with wearable devices, such as Continuous Glucose Monitoring (CGM) sensors and smartwatches, to collect real-time data on glucose levels, insulin dosages, physical activity, and sleep patterns.
-
Develop an AI Model for Hypoglycemia Prediction: Create an advanced AI model capable of predicting the risk of hypoglycemic episodes for individual patients with T1D.
-
Seamless Integration of Sensor Data: Integrate data from the "Libre 2" glucose monitoring sensor, smartwatches, and sleep tracking sensors to provide a comprehensive and real-time view of the user's health status.
-
Accurate Predictions of Nocturnal Hypoglycemic Events: Implement AI and Deep Learning algorithms to enable accurate real-time predictions, specifically focusing on nocturnal hypoglycemic events.
-
Mobile Application Development: Develop an intuitive mobile application using Flutter, delivering timely alerts, personalized insights, and educational content to users.
-
Proactive Management of Type 1 Diabetes: Ensure the system empowers individuals with T1D to proactively manage their condition through real-time alerts and user-friendly features.
-
Real-time Data Collection: Interacts with CGM sensors and smartwatches to collect real-time data on glucose levels, insulin dosages, physical activity, and sleep patterns.
-
AI-Powered Predictions: Utilizes Gated Recurrent Units (GRU) networks to predict hypoglycemic events with high accuracy.
-
User-friendly Mobile Application: Provides timely alerts, personalized insights, and educational content through an intuitive mobile application developed with Flutter.
-
Mobile Development: Flutter
-
Backend: Firebase Firestore
-
Bluetooth Low Energy (BLE) Technology: For wireless connections with CGM sensors and fitness trackers
-
AI Model Deployment: TensorFlow Lite
1. Clone the Repository:
git clone https://github.com/Youssef-Remah/Hypoglycemia_Prediction.git
2. Navigate to the Project Directory:
cd hypoglycemia-prediction
3. Install Dependencies:
flutter pub get
4. Run the Application:
flutter run
-
Connect Devices: Ensure your CGM sensor and smartwatch are connected and paired with the application.
-
Monitor Data: View real-time glucose levels, insulin dosages, and other health metrics within the app.
-
Receive Alerts: Get notified of predicted hypoglycemic events and take preventive measures.
This project is licensed under the MIT License. See the LICENSE file for details.
-
Supervised by: Dr. Ahmed Fathy Elnokrashy
-
Contributors: Youssef Remah Mohamed, Mahmoud Elrouby, Salma Ahmed Ali, Sherif Ali Mahmoud, Rawan Saeed Elnagar