Machine learning (ML) in diet planning for type-1 diabetes - An overview

Abstract

Diabetes is a chronic non-communicable disease that has become a significant public health issue for decades globally. Diabetes can slowly destroy the human body from head to leg if not appropriately treated and managed through medication, diet, and physical activities. The pancreas fails to secrete insulin or enough insulin in type-1 diabetes, while the organ secretes the same hormone sufficiently, but the body cannot process it properly in type 2. In this article, the focussed area is type-1 diabetes which is most prevalent among children and adolescents. The article tried to link up with Industry 4.0, which is a great blessing for all of us. Industries got machine learning and other applications that can help predict, analyze, assess, and intervene in diabetes and other deadly diseases. The data on type-1 diabetes can be collected from private and public settings for exploratory data analysis (EDA) followed by model selections (4ML models and Saving models). Different machine learning algorithms are usually employed for classification, prediction, and detection despite fluctuating blood sugar records. Various studies showed that an Artificial neural network (ANN) would be the best choice for these needful actions having a 34% rate of applications. In addition to these applications, calorie (diet and exercises) assessments can be done much more precisely. In conclusion, learning of machine learning has now become mandatory not only for data science people but also for physicians, dietitians, and healthcare researchers.