Pocket Dietitian: Automated Healthy Dish Recommendations by Location

Abstract

A root cause of chronic disease is a lack of timely informed decision power in everyday lifestyle choices, such as in diets. Users are unable to clearly delineate and demand healthy food in a quantitative manner. To scale the benefit of health nutrition coaching in broad real-world scenarios, we need a technological solution that is constantly able to interpret nutrition information. We ingest nutritional facts about products to efficiently calculate which items are healthiest. We deliver these results to users based on their location context. Our ranking algorithm outperforms major nutrition score metrics, and is more consistent than human dietitians in real world scenarios. Most importantly, our system gives the user a rapid way to connect with healthy food in their vicinity, reducing the barriers to a healthy diet.

Endogenous and Exogenous Multi-Modal Layers in Context Aware Recommendation Systems for Health

“…The Physio Module estimates the user physiological response to the endogenous and exogenous factors. These are mapped to the same dimensions as the nutrition features (sodium in this case) to allow for combination using the Elixir algorithm to rank foods [11]. After identifying the sodium loss from calculations below derived from literation, we get a sodium need per user-context situation (Table 4), we try to match this to the most appropriate meal within a 30km radius.…”

“…We are doing that using a combination of various algorithms in Table 1, to convert the relevant aspects of the user vector to the space of the item vectors (i.e. nutritional sodium requirements) and matching them to generate a score for how well the corresponding item satisfies the user's sodium nutritional needs [11]. As mentioned earlier, this is a great need for heart failure patients to have tailored sodium intake, yet the gold standard is a single uniform number (1500mg) given by the American Heart Association [12].…”

Live Personalized Nutrition Recommendation Engine

“…The Elixir algorithm has been previously evaluated to capture the general knowledge of food groups much better than internationally used food scoring algorithms, and with higher consistency than clinical human dietitians. Furthermore, Elixir has been shown to have the highest correlation to expert human dietitian recommendations relative to other ranking algorithms [19].…”

“…Because nutrition facts are readily available for all major restaurant chains and for packaged items, algorithms that use this information are most promising for immediate consumer use and health impact. The North American derived Nutrient Rich Foods Index 6.3 (NRF) [10], French derived SAIN/LIM method [29], and British FSA [12] all are based more heavily on available nutrition facts, yet have not been established to capture expert knowledge of dietitians or utility for individual users [19]. Current mobile applications that use nutrition facts just offer filters on the data, such as less than 600 calories [22].…”

“…We assign a personalized health score (normalized from 1–100 with 100 being healthiest) to every physically local dish and food item based on the item nutritional facts and the user parameters (which includes their daily nutritional requirements and any dietary restrictions due to preexisting medical conditions such as diabetes). Our system calculates the user parameters values as a function of real-time mobile phone sensor data and environmental parameters (Algorithm 1) [19]. Different macro nutrients are assigned a weight for calculating the score which depends on the dietary restrictions placed on or the health goals of the user.…”