Lifestyle-Based Human Health and Well Being, from Knowledge Representation to Knowledge Generation
Diet and lifestyle choices are known to generate or even revert clinical conditions like type 2 diabetes. Lifestyle engineering becomes a promising alternative to conventional approaches for health improvement. In this context, deep learning has emerged as a powerful tool to analyze observational data and enrich our knowledge about human physiology responses to diet, activity and environment. In Suggestic, we have adopted a two-fold approach to make technology work for wellbeing. First by encoding human knowledge about nutritional guidelines and making this knowledge actionable by rendering best-suited options for each individual. This encoding of actionable knowledge had several challenges, including nutritional content inference on restaurant menu items, or understanding meal composition from user natural language input. Deep learning has been used from the most basic aspects of nutritional knowledge representation to user interactions at restaurants through an Augmented Reality interface. Second by deciphering the nutritional and behavioral patterns to automatically generate knowledge for health improvement. This research is being conducted on top of established Markov Decision Process modeling tools and state of the art deep recurrent neural networks.
Ricardo holds a degree in Applied Mathematics and a Ph.D. in Biochemistry. He is Chief Data Scientist at Suggestic Inc. where he serves as co-inventor of 3 U.S. patents. His academic production has been the field of Computational Biology, specifically on the representation and analysis of protein molecular structures and novel approaches for peptide-based immunodiagnosis.