Nutrinet

NutriNet: ML-powered meal diary for diabetes management

We believe that everyone deserves to live a healthy life, regardless of their diabetes status. That's why we're committed to using data and machine learning to revolutionize diabetes management.

Our mission is to empower people with diabetes to take control of their health by providing them with personalized insights into their glucose data. We believe that everyone has the potential to live a healthy life with diabetes, and we're using data and machine learning to make that potential a reality.

Here are some specific examples of how NutriNet can help people with diabetes:

Personalized insights: NutriNet analyzes your glucose data to identify patterns and trends. This information can help you understand your diabetes better and make more informed decisions about your treatment.

Tailored recommendations: NutriNet uses your glucose data to generate personalized recommendations for diet and medication. These recommendations are based on the latest scientific research and are designed to help you achieve your individual health goals.

We believe that NutriNet is the future of diabetes care. We're excited to help people with diabetes live healthier lives and achieve their individual health goals.

Methodology

Our platform ingests raw glucose data collected using a Continuous Glucose Monitoring Device (CGM). This raw CGM data is run through our feature engineering pipeline, extracting about 100 features in total. NutriNet then uses these features to identify when users ate meals and the corresponding carbohydrate amounts per identified meal.

Impact and Use Cases

Proper diet and carb counting are essential to diabetes treatment and weight management. However, manual meal diaries are subjective and often unreliable. NutriNet provides physicians and their patients with an automated and data-driven meal diary. Users will be provided with the total number of meals and total carbohydrates per day without worrying about manually logging their meals. We believe this can provide more reliable meal data to physicians and increase the quality of diabetes management.

Acknowledgment

We would like to thank our classmates for their support and collaboration, Dr. Alberto Todeschini and Prof. Cornelia Ilin for their mentorship and guidance, and the endocrinologist and subject matter experts for their valuable insights and annotations. Your contributions were instrumental in the successful completion of this project.