Dietary information is a critical dimension for health management. This information can be used to improve our health with self-awareness. If we can monitor food intake over the long term, it could enable doctors and nutritionists to study how chronic diseases are related to dietary habits and provide food suggestions for the population. Furthermore, some special populations have special needs for food intake and we need to pay extra attention to them. In this part, we mainly care about general food nutrition and type detection for normal people, and give special attention to the infants to monitor baby food nutrients. We introduce NIRSCAM, a new design that can recognize food calories and nutrients during the intake process, without on-body instruments. Furthermore, we build a low-cost spectrometer system for accurate baby food nutrient monitoring, named BabyNutri, which costs less than 10 dollars but achieves comparable performance to commercial solutions.

Related publications:

• [ACM IMWUT/Ubicomp’23] Haiyan Hu, Qianyi Huang, and Qian Zhang. “BabyNutri: A Cost-Effective Baby Food Macronutrients Analyzer Based on Spectral Reconstruction.” [paper]

• [IEEE IoTJ’22] Haiyan Hu, Qian Zhang and Yanjiao Chen, “NIRSCam: A Mobile Near-Infrared Sensing System for Food Calorie Estimation.” [paper]