Title: Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge
With rapid growth in unhealthy diet behaviors, implementing strategies that improve healthy eating is becoming increasingly important. One approach to improving diet behavior is to continuously monitor dietary intake (e.g., calorie intake) and provide educational, motivational, and dietary recommendation feedback. Although technologies based on wearable sensors, mobile applications, and light-weight cameras exist to gather diet-related information such as food type and eating time, there remains a gap in research on how to use such information to close the loop and provide feedback to the user to improve healthy diet. We address this knowledge gap by introducing a diet behavior change framework that generates real-time diet recommendations based on a user’s food intake and considering user’s deviation from the suggested diet routine. We formulate the problem of optimal diet recommendation as a sequential decision making problem and design a greedy algorithm that provides diet recommendations such that the amount of change in user’s dietary habits is minimized while ensuring that the user’s diet goal is achieved within a given time-frame. This novel approach is inspired by the Social Cognitive Theory, which emphasizes behavioral monitoring and small incremental goals as being important to behavior change. Our optimization algorithm integrates data from a user’s past dietary intake as well as the USDA nutrition dataset to identify optimal diet changes. We demonstrate the feasibility of our optimization algorithms for diet behavior change using real-data collected in two study cohorts with a combined N=10 healthy participants who recorded their diet for up to 21 days. more »« less
Lee, Y. Q.; Collins, C. E.; Schumacher, T. L.; Weatherall, L. J.; Keogh, L.; Sutherland, K.; Gordon, A.; Rae, K. M.; Pringle, K. G.(
, Journal of Human Nutrition and Dietetics)
AbstractBackground
Little is known about the adequacy of nutrient intakes and the overall diet quality of Indigenous Australian pregnant women. The aim of this cross‐sectional study was to assess nutrient sufficiency and diet quality, as measured using the Australian Recommended Food Score (ARFS), in pregnant women from theGomeroi gaaynggalcohort (n= 58).
Methods
Maternal dietary intake during pregnancy was assessed using the Australian Eating Survey Food Frequency Questionnaire, which was self‐administered in the third trimester. Diet quality was determined using theARFS. Food group servings and nutrient intakes were compared to the Australian Guide to Health Eating (AGHE) and Australian Nutrient Reference Values (NRVs). The current analysis examined the adequacy of usual intakes from food sources only, excluding supplements.
Results
None of the women met allAGHEdaily food group serving recommendations. The highest alignment rates were for dairy (33%), meat/alternatives (31%) and vegetables (29.3%). Almost 93% of participants exceeded the recommended intake of energy‐dense, nutrient‐poor foods and percentage energy from saturated fat was high (15%). Of the five key nutrients for optimal reproductive health (folate, iron, calcium, zinc and fibre), the nutrients with the highest percentage of pregnant women achieving theNRVs were zinc (77.6%) and folate (68.9%), whereas iron was the lowest. Only one person achieved allNRVs (folate, iron, calcium, zinc and fibre) important in pregnancy. The medianARFSwas 28 points (maximum of 73).
Conclusions
Although the small cohort limits the generalisability of the findings of the present study, the data obtained indicate that the diets of these Indigenous pregnant women are inadequate. Therefore, strategies aiming to optimise nutrient intakes of Indigenous pregnant women are needed urgently.
Liu, Bulou; Bai, Bing; Xie, Weibang; Guo, Yiwen; Chen, Hao(
, ACM SIGKDD Conference on Knowledge Discovery and Data Mining)
This paper reports our recent practice of recommending articles to cold-start users at Tencent. Transferring knowledge from information-rich domains to help user modeling is an effective way to address the user-side cold-start problem. Our previous work demonstrated that general-purpose user embeddings based on mobile app usage helped article recommendations. However, high-dimensional embeddings are cumbersome for online usage, thus limiting the adoption. On the other hand, user clustering, which partitions users into several groups, can provide a lightweight, online-friendly, and explainable way to help recommendations. Effective user clustering for article recommendations based on mobile app usage faces unique challenges, including (1) the gap between an active user’s behavior of mobile app usage and article reading, and (2) the gap between mobile app usage patterns of active and cold-start users. To address the challenges, we propose a tailored Dual Alignment User Clustering (DAUC) model, which applies a sample-wise contrastive alignment to liminate the gap between active users’ mobile app usage and article reading behavior, and a distribution-wise adversarial alignment to eliminate the gap between active users’ and cold-start users’ app usage behavior. With DAUC, cold-start recommendation-optimized user clustering based on mobile app usage can be achieved. On top of the user clusters, we further build candidate generation strategies, real-time features, and corresponding ranking models without much engineering difficulty. Both online and offline experiments demonstrate the effectiveness of our work.
Perkins, Sarah; Daley, Angela; Yerxa, Kathryn; Therrien, Mona(
, American Journal of Lifestyle Medicine)
The objective of the study was to assess the relationship between participation in the Expanded Food and Nutrition Education Program (EFNEP) and food-related behavior change, as measured by the Healthy Eating Index (HEI)-2005, using a single-state, case study approach. Pre-EFNEP and post-EFNEP participation HEI scores as well as demographic, geographic, socioeconomic, and program participation characteristics from fiscal years 2013-2016 were analyzed using summary statistics and Ordinary Least Squares regressions. HEI scores were adjusted for age, sex, race, rural/urban residence, county, highest grade achieved, income, number of children, public assistance programs, and number of hours in EFNEP in the analysis. The total HEI score and several HEI subscores improved from pre-EFNEP to post-EFNEP at the 5% significance level or better. Sodium and total grains scores decreased post-EFNEP: P = .003 and P = .05, respectively. Participation in Women, Infants, and Children or the Supplemental Nutrition Assistance Program had no effect on HEI scores or changes in scores post-EFNEP. Spending less than 7 hours in the program was associated with a smaller improvement in total HEI score ( P = .05) and an increased intake of sodium ( P = .03), compared with spending 7 to 16 hours in the program. Overall, EFNEP participation was associated with improvements in diet quality.
Venable, Emily M.; Machanda, Zarin; Hagberg, Lindsey; Lucore, Jordan; Otali, Emily; Rothman, Jessica M.; Uwimbabazi, Moreen; Wrangham, Richard(
, American Journal of Physical Anthropology)
AbstractObjectives
Sodium, a vital micronutrient that is often in scarce supply for tropical herbivores, is sometimes found at high concentration in decaying wood. We tested two hypotheses for chimpanzees: first, that wood‐eating facilitates acquisition of sodium; second, that wood‐eating occurs in response to the low availability of sodium from other dietary sources.
Materials and Methods
We studied the behavior of more than 50 chimpanzees of all age‐sex classes in the Kanyawara community of Kibale National Park, Uganda. We quantified the sodium content of dietary items, including wood samples from tree species that chimpanzees consumed or did not consume. To assess variation in sodium intake, we used 7 years of data on time spent feeding on plant foods, 18 months of data on rates of food intake by adult females, and 20 years of data on meat‐eating.
Results
Major dietary sources of sodium were wood, fruits and meat. Chimpanzees consumed wood primarily from decaying trees ofNeoboutonia macrocalyx(Euphorbiaceae), which had substantially higher sodium content than all other dietary items tested. Wood‐eating was negatively correlated with fruit‐eating. Females ate wood more often than males, while males had a greater probability of consuming meat at predation events.
Discussion
We propose that females ate wood more often than males because females had reduced access to meat, their preferred source of sodium. This hypothesis suggests that the need for sodium is a motivating reason for chimpanzees to consume both meat and wood.
Automatic food type recognition is an essential task of dietary monitoring. It helps medical professionals recognize a user’s food contents, estimate the amount of energy intake, and design a personalized intervention model to prevent many chronic diseases, such as obesity and heart disease. Various wearable and mobile devices are utilized as platforms for food type recognition. However, none of them has been widely used in our daily lives and, at the same time, socially acceptable enough for continuous wear. In this paper, we propose a food type recognition method that takes advantage of Airpods Pro, a pair of widely used wireless in-ear headphones designed by Apple, to recognize 20 different types of food. As far as we know, we are the first to use this socially acceptable commercial product to recognize food types. Audio and motion sensor data are collected from Airpods Pro. Then 135 representative features are extracted and selected to construct the recognition model using the lightGBM algorithm. A real-world data collection is conducted to comprehensively evaluate the performance of the proposed method for seven human subjects. The results show that the average f1-score reaches 94.4% for the ten-fold cross- validation test and 96.0% for the self-evaluation test.
Arefeen, Asiful, Jaribi, Niloo, Mortazavi, Bobak J., and Ghasemzadeh, Hassan. Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge. Retrieved from https://par.nsf.gov/biblio/10389472. 2022 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE) .
Arefeen, Asiful, Jaribi, Niloo, Mortazavi, Bobak J., and Ghasemzadeh, Hassan.
"Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge". 2022 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE) (). Country unknown/Code not available. https://par.nsf.gov/biblio/10389472.
@article{osti_10389472,
place = {Country unknown/Code not available},
title = {Computational Framework for Sequential Diet Recommendation: Integrating Linear Optimization and Clinical Domain Knowledge},
url = {https://par.nsf.gov/biblio/10389472},
abstractNote = {With rapid growth in unhealthy diet behaviors, implementing strategies that improve healthy eating is becoming increasingly important. One approach to improving diet behavior is to continuously monitor dietary intake (e.g., calorie intake) and provide educational, motivational, and dietary recommendation feedback. Although technologies based on wearable sensors, mobile applications, and light-weight cameras exist to gather diet-related information such as food type and eating time, there remains a gap in research on how to use such information to close the loop and provide feedback to the user to improve healthy diet. We address this knowledge gap by introducing a diet behavior change framework that generates real-time diet recommendations based on a user’s food intake and considering user’s deviation from the suggested diet routine. We formulate the problem of optimal diet recommendation as a sequential decision making problem and design a greedy algorithm that provides diet recommendations such that the amount of change in user’s dietary habits is minimized while ensuring that the user’s diet goal is achieved within a given time-frame. This novel approach is inspired by the Social Cognitive Theory, which emphasizes behavioral monitoring and small incremental goals as being important to behavior change. Our optimization algorithm integrates data from a user’s past dietary intake as well as the USDA nutrition dataset to identify optimal diet changes. We demonstrate the feasibility of our optimization algorithms for diet behavior change using real-data collected in two study cohorts with a combined N=10 healthy participants who recorded their diet for up to 21 days.},
journal = {2022 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE)},
author = {Arefeen, Asiful and Jaribi, Niloo and Mortazavi, Bobak J. and Ghasemzadeh, Hassan},
}
Warning: Leaving National Science Foundation Website
You are now leaving the National Science Foundation website to go to a non-government website.
Website:
NSF takes no responsibility for and exercises no control over the views expressed or the accuracy of
the information contained on this site. Also be aware that NSF's privacy policy does not apply to this site.
