Abstract

The proportion of people aged 60 years or older is growing faster than ever and expected to double from 2015 to 2050. Mainly due to a higher life expectancy and decrease in birth rates, this poses challenges especially for health care facilities like retirement homes. Because problems in the nutrition of individuals are often the cause or related to health problems of seniors, ensuring the correct nutrition and thus preventing a lot of health impairments would be a very effective way of improving the health of seniors and lessen the workload on caregivers and nurses. In this thesis, we develop a mobile application to help caregivers track and analyze the nutritional intake of their patients. Using a smart weighing gadget, meals are weighed before and after each patient consumes them, with the ability to specify the distribution of different dishes enabling the app to accurately calculate intakes for specific nutrients. Theses intakes can then be compared against predefined recommendations tailored to the individual requirements of each patient, even respecting eventual diseases altering the nutritional needs. This information is then used to predict future nutrient intakes with the goal of recognizing incipient deficiencies or excess consumption before they actually occur. To be able to achieve this, different forecasting algorithms have been evaluated with Holt's linear trend method performing best for this use case. Together with other tools like recording and interpreting the weight history of each patient, caretakers are provided with a powerful tool to be able to identify and prevent malnutrition more easily.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]