Importance of Food Recognition on Blood Glucose Monitoring
Disciplines
Artificial Intelligence and Robotics | Dietetics and Clinical Nutrition | Software Engineering
Abstract (300 words maximum)
Diabetes has emerged as a worldwide health issue even when over 50% of type 2 diabetes cases are preventable. Maintaining blood sugar under control requires eating a healthy and balanced diet, exercising, and adhering to medications. Dietary consumption must be under strict control for diabetic patients’ general health. Traditional techniques for monitoring dietary consumption include recollection and manual record-keeping, but they can be tedious and prone to mistakes when used repeatedly. However, automated technologies for maintaining records that make use of computer vision and mobile cameras, such as food image recognition systems (FIRS), can streamline the process and help diabetes patients better manage their chronic health condition by automating their diet tracking. These solutions seek to efficiently track daily food intake and then offer nutritional suggestions to facilitate and encourage lifestyle improvements. Thus, in this work, we are designing and implementing a Machine Learning model that can recognize/classify food categories and estimate the corresponding volume and calorific content from picture(s) of an upcoming meal, which would help users assess the effect of the intake on their blood sugar levels. This is part of a larger project that involves an application to help GlucoCheck — a non-invasive blood glucose monitoring device — users keep track of their blood glucose levels and possible spikes. The majority of research exclusively concentrate on calorie estimation without making any direct connection to diabetic patients’ blood sugar levels. The main difference between our model and other similar ones is its direct application to diabetes management.
Academic department under which the project should be listed
CCSE - Computer Science
Primary Investigator (PI) Name
Maria Valero de Clemente
Importance of Food Recognition on Blood Glucose Monitoring
Diabetes has emerged as a worldwide health issue even when over 50% of type 2 diabetes cases are preventable. Maintaining blood sugar under control requires eating a healthy and balanced diet, exercising, and adhering to medications. Dietary consumption must be under strict control for diabetic patients’ general health. Traditional techniques for monitoring dietary consumption include recollection and manual record-keeping, but they can be tedious and prone to mistakes when used repeatedly. However, automated technologies for maintaining records that make use of computer vision and mobile cameras, such as food image recognition systems (FIRS), can streamline the process and help diabetes patients better manage their chronic health condition by automating their diet tracking. These solutions seek to efficiently track daily food intake and then offer nutritional suggestions to facilitate and encourage lifestyle improvements. Thus, in this work, we are designing and implementing a Machine Learning model that can recognize/classify food categories and estimate the corresponding volume and calorific content from picture(s) of an upcoming meal, which would help users assess the effect of the intake on their blood sugar levels. This is part of a larger project that involves an application to help GlucoCheck — a non-invasive blood glucose monitoring device — users keep track of their blood glucose levels and possible spikes. The majority of research exclusively concentrate on calorie estimation without making any direct connection to diabetic patients’ blood sugar levels. The main difference between our model and other similar ones is its direct application to diabetes management.