Short Paper: Use of Natural Spoken Language with Automated Mapping of Self-Reported Food Intake to Food Composition Data for Low-Burden Real-Time Dietary Assessment (Preprint)

2021 ◽  
Author(s):  
Salima Taylor ◽  
Mandy Korpuski ◽  
Sai Das ◽  
Cheryl Gilhooly ◽  
Ryan Simpson ◽  
...  
Keyword(s):  
Food Intake ◽  
Dietary Intake ◽  
Energy Intake ◽  
Dietary Assessment ◽  
Spoken Language ◽  
Assessment Method ◽  
Food Composition ◽  
Short Paper ◽  
Self Monitoring ◽  
Significant Difference

BACKGROUND Self-monitoring food intake is a cornerstone of national recommendations for health, but existing applications are burdensome, which limits use. OBJECTIVE We developed and pilot tested a new app (COCO Nutritionist) that combines speech understanding technology with technologies for mapping foods to appropriate food composition codes in national databases, for lower-burden and automated nutritional analysis of self-reported dietary intake. METHODS COCO was compared with the multiple-pass, interviewer-administered 24h-recall method for assessment of energy intake. COCO was used for five consecutive days, and 24-h dietary recalls were obtained for two of the days. Participants were 35 women and men with a mean age of 28 (range 20-58) years, and mean Body Mass Index of 24 (range 17-48) kg/m2. RESULTS There was no significant difference in energy intake between values obtained by COCO and 24-h recall for days when both methods were used (2092 +/- 1044 [SD] versus 2030 +/- 687 [SD], P=0.70). There was also no differences between the methods in the percent of energy from protein, carbohydrate and fat (P=0.27-0.89), and no trend in energy intake obtained with COCO over the entire 5-day study period (p=0.186). CONCLUSIONS This first demonstration of a dietary assessment method using natural spoken language to map reported foods to food composition codes demonstrates a promising new approach to automate assessments of dietary intake. CLINICALTRIAL N/A

scholarly journals Testing the Validity of a Natural Spoken Language Application for the Self-Monitoring of Daily Dietary Intake (P13-035-19)

2019 ◽  
Vol 3 (Supplement_1) ◽  
Author(s):  
Rachel Silver ◽  
Mandy Korpusik ◽  
Salima Taylor ◽  
Sai Das ◽  
Cheryl Gilhooly ◽  
...  
Keyword(s):  
Pilot Study ◽  
Dietary Intake ◽  
Energy Intake ◽  
Reference Method ◽  
Spoken Language ◽  
Self Monitoring ◽  
Daily Dietary Intake ◽  
Dietary Recall ◽  
Language Technology ◽  
Significant Difference

Abstract Objectives Self-monitoring daily dietary intake is recommended for weight loss, weight maintenance, and healthy eating. However, current tracking methods are often burdensome and result in short-term use. We conducted a pilot study to evaluate the accuracy of a new application designed to self-monitor dietary intake using natural spoken language (COCO; The Conversational Calorie Counter). Methods A total of 14 participants were recruited for the pilot study. They were instructed to record daily dietary intake using the COCO application for at least five consecutive days. Two unscheduled 24-hour dietary recalls were conducted between day 3 and day 5 as the reference method for evaluating total energy intake (TEI). The two-day energy estimates were averaged for each assessment method. Pearson's correlation coefficient was used to assess the validity of the COCO application. Estimates of TEI from COCO were compared to the 24-hour dietary recall by a paired samples t-test. Results Participants were primarily female (86%), with an average body mass index of 22.2 ± 1.8 kg/m2 (mean ± standard deviation). On average, participants consumed three daily meals and recorded dietary intake for six days using the COCO application. The average TEI was 1782 ± 773 kcal for all recorded days (range: 4 to 10). The mean TEI measured by 24-hour dietary recall was 1791 ± 862 kcal, and mean TEI measured by COCO for the corresponding days was 1818 ± 916 kcal. We observed a significant correlation between the assessment methods (r = 0.58; P = 0.03), and there was no significant difference in TEI estimates from COCO compared to the 24-hour recall (P = 0.90). Conclusions These results suggest that natural spoken language technology can be used in applications that facilitate self-monitoring of food intake to support weight management and the prevention of noncommunicable diseases. The significant correlation between estimates of TEI from COCO and the 24-hour dietary recall indicates the potential validity of this novel approach for capturing dietary data and assessing energy intake. Funding Sources Sponsored by the National Institutes of Health (R21HL118347), the U.S. Department of Agriculture with Tufts University (58–1950-4–003), Quanta Computing, Inc., and the Department of Defense (National Defense Science Engineering Graduate Fellowship Program).

scholarly journals Comparison of Energy Intake Determined by a Natural Spoken Language Application with 24-h Recall

2020 ◽  
Vol 4 (Supplement_2) ◽  
pp. 1694-1694
Author(s):  
Salima Taylor ◽  
Mandy Korpusik ◽  
Rachel Silver ◽  
Sai Krupa Das ◽  
Cheryl Gilhooly ◽  
...  
Keyword(s):  
Weight Loss ◽  
Pilot Study ◽  
Food Intake ◽  
Dietary Intake ◽  
Energy Intake ◽  
Reference Method ◽  
Spoken Language ◽  
Assessment Method ◽  
Daily Dietary Intake ◽  
Language Technology

Abstract Objectives Self-monitoring daily dietary intake is recommended for weight loss and weight loss maintenance. However, current online platforms and applications are often burdensome, which may limit use. We conducted a pilot study to evaluate the accuracy of a new application designed to self-monitor dietary intake using natural spoken language (COCO; The Conversational Calorie Counter). Methods A total of 35 participants were enrolled in this pilot study. Participants were asked to record daily dietary intake using the COCO application for a period of at least five days. Two 24-hour dietary recalls were conducted during this time, between day three and day five, and served as the reference method for evaluating total energy intake (TEI; measured in kcal). Mean two-day energy intake was calculated for each assessment method for the days when the 24-hr recall and COCO data were collected. Self-reported TEI from COCO were compared to estimates obtained from the 24-hour dietary recalls by a paired samples t-test and a Pearson's correlation coefficient. Results On average, participants consumed three meals a day and recorded six days of food intake days with COCO (range: 4 to 10 days). The mean TEI was not significantly different between the two methods (1902 ± 621 kcal by 24-hour dietary recall and 1988 ± 1033 kcal by COCO, P = 0.59). There was a significant correlation between mean TEI measured with the two methods (r = 0.45; P = 0.006). In addition, a strong correlation was observed between the number of food items logged in COCO and those recalled in the 24-hour diet recalls (r = 0.82; P >0.0001). Completion of the exit survey by 28 participants indicated that 43% would definitely or probably use the application again. Conclusions These results suggest that natural spoken language technology may have utility in applications to self-monitor food intake. Additional research is required to fully elucidate the validity of COCO in estimating dietary intake. Funding Sources This research was supported by the NIH Grant # 1R21HL118347–01 (SBR and JG), Quanta Computing, Inc., and the National Defense Science and Engineering Graduate fellowship.

scholarly journals Methodology for Objective, Passive, Image- and Sensor-based Assessment of Dietary Intake, Meal-timing, and Food-related Activity in Ghana and Kenya (P13-028-19)

2019 ◽  
Vol 3 (Supplement_1) ◽  
Author(s):  
Megan McCrory ◽  
Mingui Sun ◽  
Edward Sazonov ◽  
Gary Frost ◽  
Alex Anderson ◽  
...  
Keyword(s):  
Food Intake ◽  
Dietary Intake ◽  
Dietary Assessment ◽  
Food Composition ◽  
Primary Caregiver ◽  
Middle Income ◽  
Food Shopping ◽  
Meal Timing ◽  
Custom Software ◽  
Wearable Cameras

Abstract Objectives Herein we describe a new system we have developed for assessment of dietary intake, meal timing, and food-related activities, adapted for use in low- and middle-income countries. Methods System components include one or more wearable cameras (the Automatic Ingestion Monitor-2 (AIM), an eyeglasses-mounted wearable chewing sensor and micro-camera; ear-worn camera; the eButton, a camera attached to clothes; and eHat, a camera attached to a visor worn by the mother when feeding infants and toddlers), and custom software for evaluation of dietary intake from food-based images and sensor-detected food intake. General protocol: The primary caregiver of the family uses one or more wearable cameras during all waking hours. The cameras aim directly in front of the participant and capture images every few seconds, thereby providing multiple images of all food-related activities throughout the day. The camera may be temporarily removed for short periods to preserve privacy, such as during bathing and personal care. For analysis, images and sensor signals are processed by the study team in custom software. The images are time-stamped, arranged in chronological order, and linked with sensor-detected eating occasions. The software also incorporates food composition databases of choice such as the West African Foods Database, a Kenyan Foods Database, and the USDA Food Composition Database, allowing for image-based dietary assessment by trained nutritionists. Images can be linked with nutritional analysis and tagged with an activity label (e.g., food shopping, child feeding, cooking, eating). Assessment of food-related activities such as food-shopping, food gathering from gardens, cooking, and feeding of other family members by the primary caregiver can help provide context for dietary intake and additional information to increase accuracy of dietary assessment and analysis of eating behavior. Examples of the latter include assessment of specific ingredients in prepared dishes, the source of these ingredients, cooking method, and how, where, and when food is consumed. Results N/A. Conclusions Pilot- and feasibility-testing is underway. The system will be tested for accuracy of dietary intake assessment versus weighed food intake in urban and rural settings around Accra, Ghana and Nairobi, Kenya. Funding Sources [Funded by the Bill & Melinda Gates Foundation].

A comparison of the Remote Food Photography Method and the Automated Self-Administered 24-hour Dietary Assessment Tool for measuring full day dietary intake among school-aged children

2021 ◽  
pp. 1-26
Author(s):  
Traci A. Bekelman ◽  
Corby K. Martin ◽  
Susan L. Johnson ◽  
Deborah H. Glueck ◽  
Katherine A. Sauder ◽  
...  
Keyword(s):  
Dietary Intake ◽  
Energy Intake ◽  
Mixed Model ◽  
Linear Mixed Model ◽  
Assessment Tool ◽  
Dietary Assessment ◽  
Ease Of Use ◽  
Self Report ◽  
Food Photography ◽  
Photography Method

Abstract The limitations of self-report measures of dietary intake are well known. Novel, technology-based measures of dietary intake may provide a more accurate, less burdensome alternative to existing tools. The first objective of this study was to compare participant burden for two technology-based measures of dietary intake among school-age children: the Automated-Self Administered 24-hour Dietary Assessment Tool-2018 (ASA24-2018) and the Remote Food Photography Method (RFPM). The second objective was to compare reported energy intake for each method to the Estimated Energy Requirement for each child, as a benchmark for actual intake. Forty parent-child dyads participated in 2, 3-day dietary assessments: a parent proxy-reported version of the ASA24 and the RFPM. A parent survey was subsequently administered to compare satisfaction, ease of use and burden with each method. A linear mixed model examined differences in total daily energy intake (TDEI) between assessments, and between each assessment method and the EER. Reported energy intake was 379 kcal higher with the ASA24 than the RFPM (p=0.0002). Reported energy intake with the ASA24 was 231 kcal higher than the EER (p = 0.008). Reported energy intake with the RFPM did not differ significantly from the EER (difference in predicted means = −148 kcal, p = 0.09). Median satisfaction and ease of use scores were 5 out of 6 for both methods. A higher proportion of parents reported that the ASA24 was more time consuming than the RFPM (74.4% vs. 25.6%, p = 0.002). Utilization of both methods is warranted given their high satisfaction among parents.

scholarly journals Effect of Manual Data Cleaning on Nutrient Intakes using the Automated Self-Administered 24-Hour Dietary Assessment Tool (ASA24)

2021 ◽  
Author(s):  
Yasmine Y Bouzid ◽  
Joanne E Arsenault ◽  
Ellen L Bonnel ◽  
Eduardo Cervantes ◽  
Annie Kan ◽  
...  
Keyword(s):  
Energy Intake ◽  
Assessment Tool ◽  
Data Cleaning ◽  
Dietary Assessment ◽  
Assessment Tools ◽  
Energy Requirements ◽  
Nutrient Intakes ◽  
Significant Difference ◽  
Before And After ◽  
Total Fat

Abstract Background Automated dietary assessment tools such as ASA24® are useful for collecting 24-hour recall data in large-scale studies. Modifications made during manual data cleaning may affect nutrient intakes. Objectives We evaluated the effects of modifications made during manual data cleaning on nutrients intakes of interest: energy, carbohydrate, total fat, protein, and fiber. Methods Differences in mean intake before and after data cleaning modifications for all recalls and average intakes per subject were analyzed by paired t-tests. Chi-squared test was used to determine whether unsupervised recalls had more open-ended text responses that required modification than supervised recalls. We characterized food types of text response modifications. Correlations between predictive energy requirements, measured total energy expenditure (TEE), and mean energy intake from raw and modified data were examined. Results After excluding 11 recalls with invalidating technical errors, 1499 valid recalls completed by 393 subjects were included in this analysis. We found significant differences before and after modifications for energy, carbohydrate, total fat, and protein intakes for all recalls (p < 0.05). Limiting to modified recalls, there were significant differences for all nutrients of interest, including fiber (p < 0.02). There was not a significantly greater proportion of text responses requiring modification for home compared to supervised recalls (p = 0.271). Predicted energy requirements correlated highly with TEE. There was no significant difference in correlation of mean energy intake with TEE for modified compared to raw data. Mean intake for individual subjects was significantly different for energy, protein, and fat intakes following cleaning modifications (p < 0.001). Conclusions Manual modifications can change mean nutrient intakes for an entire cohort and individuals. However, modifications did not significantly affect correlation of energy intake with predictive requirements and measured expenditure. Investigators can consider their research question and nutrients of interest when deciding to make cleaning modifications.

Dietary Assessment and Self-monitoring: With Nutrition Applications for Mobile Devices

2012 ◽  
Vol 73 (3) ◽  
pp. e253-e260 ◽  
Author(s):  
Jessica R. L. Lieffers ◽  
Rhona M. Hanning
Keyword(s):  
Dietary Intake ◽  
Mobile Devices ◽  
Dietary Assessment ◽  
Portion Size ◽  
Personal Digital Assistants ◽  
Self Monitoring ◽  
Food Records ◽  
Exciting Potential ◽  
Conventional Methods ◽  
24 Hour Recall

Nutrition applications for mobile devices (e.g., personal digital assistants, smartphones) are becoming increasingly accessible and can assist with the difficult task of intake recording for dietary assessment and self-monitoring. This review is a compilation and discussion of research on this tool for dietary intake documentation in healthy populations and those trying to lose weight. The purpose is to compare this tool with conventional methods (e.g., 24-hour recall interviews, paperbased food records). Research databases were searched from January 2000 to April 2011, with the following criteria: healthy or weight loss populations, use of a mobile device nutrition application, and inclusion of at least one of three measures, which were the ability to capture dietary intake in comparison with conventional methods, dietary self-monitoring adherence, and changes in anthropometrics and/or dietary intake. Eighteen studies are discussed. Two application categories were identified: those with which users select food and portion size from databases and those with which users photograph their food. Overall, positive feedback was reported with applications. Both application types had moderate to good correlations for assessing energy and nutrient intakes in comparison with conventional methods. For self-monitoring, applications versus conventional techniques (often paper records) frequently resulted in better self-monitoring adherence, and changes in dietary intake and/or anthropometrics. Nutrition applications for mobile devices have an exciting potential for use in dietetic practice.

scholarly journals Food Composition Impacts the Accuracy of Wearable Devices When Estimating Energy Intake from Energy-Dense Food

Nutrients ◽  
2019 ◽  
Vol 11 (5) ◽  
pp. 1170
Author(s):  
Giulia Lorenzoni ◽  
Daniele Bottigliengo ◽  
Danila Azzolina ◽  
Dario Gregori
Keyword(s):  
Food Intake ◽  
Energy Intake ◽  
Caloric Intake ◽  
Wearable Devices ◽  
Food Composition ◽  
Dominant Hand ◽  
Measurement Device ◽  
The Difference ◽  
The University ◽  
Energy Dense Food

The present study aimed to assess the feasibility and reliability of an a3utomatic food intake measurement device in estimating energy intake from energy-dense foods. Eighteen volunteers aged 20–36 years were recruited from the University of Padova. The device used in the present study was the Bite Counter (Bite Technologies, Pendleton, USA). The rationale of the device is that the wrist movements occurring in the act of bringing food to the mouth present unique patterns that are recognized and recorded by the Bite Counter. Subjects were asked to wear the Bite Counter on the wrist of the dominant hand, to turn the device on before the first bite and to turn it off once he or she finished his or her meal. The accuracy of caloric intake was significantly different among the methods used. In addition, the device’s accuracy in estimating energy intake varied according to the type and amount of macronutrients present, and the difference was independent of the number of bites recorded. Further research is needed to overcome the current limitations of wearable devices in estimating caloric intake, which is not independent of the food being eaten.

scholarly journals Dietary Intake and Clinical Response of Hospitalized Patients with Acute Diarrhea

2008 ◽  
Vol 29 (1) ◽  
pp. 25-31 ◽  
Author(s):  
Mahfuza Islam ◽  
S.K. Roy ◽  
Muktara Begum ◽  
M. Jobayer Chisti
Keyword(s):  
Food Intake ◽  
Dietary Intake ◽  
Clinical Response ◽  
Energy Intake ◽  
Acute Diarrhea ◽  
Lower Energy ◽  
Recovery Period ◽  
Severe Illness ◽  
Stool Weight ◽  
Clinical Records

Background Diarrhea and malnutrition remain major health problems among children of developing countries. During diarrhea, the patient's dietary intake and absorption of nutrients are reduced while nutritional requirements are increased. Objective To determine the relationship between food intake and clinical response during the hospital stay of patients with acute diarrhea. Methods A hospital-based longitudinal study was conducted in 118 patients with acute diarrhea aged 6 to 59 months who required treatment for at least 3 days in the in-patient ward in Dhaka Hospital of the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR, B). Daily food intake was measured and anthropometric measurements were taken to assess nutritional status. Daily stool weight and clinical records were collected. The data were analyzed with SPSS/PC+, version 10, and EPI STAT, version 3.2.2. Results The duration of diarrhea was 50% greater in patients with lower energy intake (less than 50% of the recommended dietary allowance [RDA]) than in those with higher energy intake (6 vs. 4 days, p = <.001). Patients with lower energy intake had 22% greater stool output than those with higher energy intake (122.65 vs. 100.37 mL/kg body weight/day, p = .04). Among patients with lower energy intake, the weight-for-age and weight-for-height z-scores (WAZ and WHZ) at discharge from the hospital were higher than those at admission (−3.53±1.25 vs. −3.67±1.31 and 1.95±1.23 vs. −2.14±1.22, respectively; p = .001 for both comparisons), but these scores did not differ at admission and discharge among patients with higher energy intake. The Kaplan–Meier survival function showed that 80% of well-nourished children (WAZ ≥ −2), as compared with 58% of malnourished children (WAZ < −2), recovered by the 4th day of treatment ( p < .01). The length of the recovery period was related negatively with total energy intake ( p = <.001) and mid-upper-arm circumference ( p = .004) and positively with stool weight. Conclusions Food intake was reduced in the hospitalized children because of severe illness. Patients with lower energy intake as a percentaqe of RDA had delayed clinical recovery and higher stool output.

scholarly journals Assessment of Dietary Intake Among South Asian Adults in the United States

2021 ◽  
Vol 5 (Supplement_2) ◽  
pp. 124-124
Author(s):  
Nurgul Fitzgerald ◽  
Shailja Mathur
Keyword(s):  
United States ◽  
Dietary Intake ◽  
Energy Intake ◽  
South Asian ◽  
Assessment Tool ◽  
Dietary Assessment ◽  
The United States ◽  
Management Program ◽  
Convenience Sample ◽  
Total Fat

Abstract Objectives To examine the dietary intake patterns of South Asian adults by using three different assessment methods. Methods The participants were a convenience sample of 62 adults from South Asian descent, who lived in the United States and participated in a community-based diabetes self-management program. Dietary intake data were collected through Automated Self-Administered 24-Hour Dietary Assessment Tool (ASA24), self-administered Diet History Questionnaire III (self-DHQ), and researcher-administered DHQ III (res-DHQ) (National Cancer Institute). Thirty-seven participants completed ASA24 and self-DHQ back-to-back during in-person sessions, and 25 participants completed res-DHQ through video conferencing sessions with the researcher. Group level data were examined using IBM SPSS Statistics software. Results On average, participants’ daily energy intake levels were estimated to be 805.8 ± 551.3, 1686.4 ± 985.9 and 1469.7 ± 887.5 kcal/d by self-DHQ, ASA24, and res-DHQ, respectively. Self-DHQ produced the lowest of the estimates (mean ± SD) for daily protein (28.9 ± 18.8 vs 63.1 ± 35.2, and 53.1 ± 27.9 g/d), carbohydrate (106.4 ± 68.0 vs 224.9 ± 128.4 and 199.9 ± 119.7 g/d), and total fat (31.7 ± 29.2 vs. 63.5 ± 46.5 and 56.2 ± 40.9 g/d) intakes in comparison to ASA24 and res-DHQ, respectively. Conclusions In this study, self-administered DHQ produced substantially lower estimates of daily macronutrient and energy intake levels. The ASA24 or researcher-administered DHQ were relatively more reliable methods of dietary assessment in this sample of South Asian adults. Funding Sources NJ Department of Health, Office of Minority and Multicultural Health.

Sign in / Sign up

Export Citation Format

Share Document