New mobile methods for dietary assessment: review of image-assisted and image-based dietary assessment methods

The Future of Obesity Management through Precision Nutrition: Putting the Individual at the Center

PURPOSE OF REVIEW: The prevalence of obesity continues to rise steadily. While obesity management typically relies on dietary and lifestyle modifications, individual responses to these interventions vary widely. Clinical guidelines for overweight and obesity stress the importance of personalized approaches to care. This review aims to underscore the role of precision nutrition in delivering tailored interventions for obesity management. RECENT FINDINGS: Recent technological strides have expanded our ability to detect obesity-related genetic polymorphisms, with machine learning algorithms proving pivotal in analyzing intricate genomic data. Machine learning algorithms can also predict postprandial glucose, triglyceride, and insulin levels, facilitating customized dietary interventions and ultimately leading to successful weight loss. Additionally, given that adherence to dietary recommendations is one of the key predictors of weight loss success, employing more objective methods for dietary assessment and monitoring can enhance sustained long-term compliance. Biomarkers of food intake hold promise for a more objective dietary assessment. Acknowledging the multifaceted nature of obesity, precision nutrition stands poised to transform obesity management by tailoring dietary interventions to individuals' genetic backgrounds, gut microbiota, metabolic profiles, and behavioral patterns. However, there is insufficient evidence demonstrating the superiority of precision nutrition over traditional dietary recommendations. The integration of precision nutrition into routine clinical practice requires further validation through randomized controlled trials and the accumulation of a larger body of evidence to strengthen its foundation.

Recommendations for Designing, Conducting, and Reporting Feeding Trials in Nutrition Research

Double-blind, placebo-controlled, randomized controlled trials are the gold standard for clinical trials in nutrition science. For trials of whole diets, dietary counseling is advantageous as they offer clinical translatability although can vary in the fidelity of the intended intervention from participant to participant and across studies. Feeding trials, in which most or all food is provided, offer high precision and can provide proof-of-concept evidence that a dietary intervention is efficacious and can also better evaluate the effect of known quantities of foods and nutrients on physiology. However, they come with additional methodological complexities. Feeding trials also call for a variety of unique methodological considerations, not least of which relate to the design and delivery of diets to participants. This review aims to provide a comprehensive summary of recommendations for design and conduct of feeding trials, encompassing domiciled and nondomiciled feeding trials. Several pertinent aspects of trial design and methodology are discussed, including defining the study population to maximize retention, safety, and generalizability of findings, recommendations for design of control interventions and optimizing blinding, and specific considerations for clinical populations. A detailed stepwise process for menu design, development, validation, and delivery are also presented. These recommendations aim to facilitate methodologic consistency and execution of high-quality feeding trials, ultimately facilitating improved understanding of the role of diet in treating disease and the underpinning mechanisms.

Evaluating the Quality and Comparative Validity of Manual Food Logging and Artificial Intelligence-Enabled Food Image Recognition in Apps for Nutrition Care

For artificial intelligence (AI) to support nutrition care, high quality and accuracy of its features within smartphone applications (apps) are essential. This study evaluated popular apps' features, quality, behaviour change potential, and comparative validity of dietary assessment via manual logging and AI. The top 200 free and paid nutrition-related apps from Australia's Apple App and Google Play stores were screened (n = 800). Apps were assessed using MARS (quality) and ABACUS (behaviour change potential). Nutritional outputs from manual food logging and AI-enabled food-image recognition apps were compared with food records for Western, Asian, and Recommended diets. Among 18 apps, Noom scored highest on MARS (mean = 4.44) and ABACUS (21/21). From 16 manual food-logging apps, energy was overestimated for Western (mean: 1040 kJ) but underestimated for Asian (mean: -1520 kJ) diets. MyFitnessPal and Fastic had the highest accuracy (97% and 92%, respectively) out of seven AI-enabled food image recognition apps. Apps with more AI integration demonstrated better functionality, but automatic energy estimations from AI-enabled food image recognition were inaccurate. To enhance the integration of apps into nutrition care, collaborating with dietitians is essential for improving their credibility and comparative validity by expanding food databases. Moreover, training AI models are needed to improve AI-enabled food recognition, especially for mixed dishes and culturally diverse foods.

Measures of nutrition status and health for weight-inclusive patient care: A narrative review

In healthcare, weight is often equated to and used as a marker for health. In examining nutrition and health status, there are many more effective markers independent of weight. In this article, we review practical and emerging clinical applications of technologies and tools used to collect non-weight-related data in nutrition assessment, monitoring, and evaluation in the outpatient setting. The aim is to provide clinicians with new ideas about various types of data to evaluate and track in nutrition care.

Validity of the PortionSize application compared with that of MyFitnessPal for accurately estimating intake: a randomized crossover laboratory-based evaluation

BACKGROUND

PortionSize offers real-time feedback on dietary intake, including intake of MyPlate food groups but requires further evaluation on a larger sample in a laboratory-based setting. MyFitnessPal (MFP) is a commonly used commercial dietary assessment application, and to our knowledge, no known studies have evaluated MFP in a laboratory setting.

OBJECTIVES

The overall objective was to test the validity of PortionSize and MFP to accurately measure intake compared with that of weighed food (WB) and to compare error between applications. A secondary objective was to test usability, satisfaction, and user preference between applications.

METHODS

This randomized crossover study was completed between February and October 2021. Participants (N = 43) used both applications to estimate intake in a laboratory setting. Participants were provided with a preweighed plated meal and plated leftovers. Two 1-sided t tests assessed equivalence (±21% bounds) between simulated intake from PortionSize and WB, and MFP and WB. The primary outcome was energy intake, and secondary outcome measures were portion size (in grams), food groups, and other nutrients. Differences in relative absolute error, usability, satisfaction, and user preference between applications were evaluated using dependent samples t tests. Cohen d assessed effect size.

RESULTS

For PortionSize, energy and portion size were underestimated by 13.3% and 14.0%, respectively, and were not equivalent to WB. For MFP, energy was overestimated by 7.0%, and equivalent to WB (P = 0.04). Relative absolute error for energy did not differ between applications. For PortionSize, Cohen d was small (<0.2) for fruits, grains, protein foods, and specific nutrients. No differences were seen with usability, and the only difference for satisfaction was that participants found it easier to use MFP to find foods consumed (P = 0.019), and participants preferred using MFP (P = 0.014).

CONCLUSIONS

PortionSize requires further updates to improve energy estimates and usability but demonstrates clinical utility for tracking food group and nutrient intake. PortionSize did not outperform MFP for measuring energy intake.

CLINICAL TRIAL REGISTRY

This trial was registered at clinicaltrials.gov as NCT04700904 (https://classic.

CLINICALTRIALS

gov/ct2/show/NCT04700904).

Validity of a digital photo-based dietary assessment tool: Development and initial evaluation

Background and aim: To evaluate the validity and user satisfaction of a digital photo-based dietary assessment tool as an alternative to the hand-written paper record method that assists researchers during the pandemic. This study compared nutrient intake and users' satisfaction with methods between a digital photo-based dietary assessment tool, known as the Nutrinote Gama app, and food weighing as the gold standard. Methods: Fifty college students majoring in food and nutrition (90% were women; median age, 21 years) took pictures of their foods and beverages before and after consumption and then uploaded them to the Nutrinote Gama application. Trained nutritionists evaluated plate wastes, and nutritional content was revealed on the Nutrinote Gama application. Parallel to the photo-based method, they kept a weight dietary record and sent it to the researcher. A questionnaire was used to assess participants' satisfaction. Results: No statistical differences (p = 0.89) were observed in the measurement of energy intake between Nutrinote Gama (mean ± standard deviation [SD] = 582.8 ± 131) and food weighing (mean ± SD = 566.1 ± 133). No statistical differences (p = 0.59) were also observed in the measurement of protein, fat (p = 0.434), and carbohydrate (p = 230). The energy, protein, fat, and carbohydrate intakes estimated from the two methods were significantly correlated (r = 0.86, 0.870, 0.811, 0.738, respectively). Over 70% of participants were satisfied with the photo-based record. Conclusion: The results indicate that this digital photo-based dietary assessment tool is valid and user-friendly to estimate nutrient intake.

An Overview of Traditional and Novel Tools to Assess Diet

Effective dietary interventions are vital for combating morbidity and mortality, necessitating reliable assessment tools. This article explores diverse dietary assessment methods, emphasizing their complexities and applications. Food Frequency Questionnaires (FFQs) offer insights into dietary habits over specified periods but require validation for target populations. Traditional Food Records provide detailed insights but are labor-intensive and prone to underreporting. Technology-based and technology-assisted records offer efficient alternatives, leveraging mobile apps and wearable sensors, albeit with access and privacy concerns. 24 hour Dietary Recall (24HR) methods capture detailed intake within a day, with traditional and technology-assisted approaches strengthening population studies. The Automated Multi-Pass Method (AMPM) and technology-assisted ASA24 represent traditional and contemporary 24HR methodologies, respectively, both yielding comprehensive dietary data. In conclusion, dietary assessments are crucial for understanding dietary patterns and health implications. Integration of novel technologies streamlines data collection and analysis, enhancing researchers' ability to accurately gauge short- and long-term dietary impacts. Proper utilization of these tools empowers researchers to make informed decisions regarding dietary interventions and public health initiatives.

Accuracy of energy and nutrient intake estimation versus observed intake using 4 technology-assisted dietary assessment methods: a randomized crossover feeding study

BACKGROUND

Technology-assisted 24-h dietary recalls (24HRs) have been widely adopted in population nutrition surveillance. Evaluations of 24HRs inform improvements, but direct comparisons of 24HR methods for accuracy in reference to a measure of true intake are rarely undertaken in a single study population.

OBJECTIVES

To compare the accuracy of energy and nutrient intake estimation of 4 technology-assisted dietary assessment methods relative to true intake across breakfast, lunch, and dinner.

METHODS

In a controlled feeding study with a crossover design, 152 participants [55% women; mean age 32 y, standard deviation (SD) 11; mean body mass index 26 kg/m2, SD 5] were randomized to 1 of 3 separate feeding days to consume breakfast, lunch, and dinner, with unobtrusive weighing of foods and beverages consumed. Participants undertook a 24HR the following day [Automated Self-Administered Dietary Assessment Tool-Australia (ASA24); Intake24-Australia; mobile Food Record-Trained Analyst (mFR-TA); or Image-Assisted Interviewer-Administered 24-hour recall (IA-24HR)]. When assigned to IA-24HR, participants referred to images captured of their meals using the mobile Food Record (mFR) app. True and estimated energy and nutrient intakes were compared, and differences among methods were assessed using linear mixed models.

RESULTS

The mean difference between true and estimated energy intake as a percentage of true intake was 5.4% (95% CI: 0.6, 10.2%) using ASA24, 1.7% (95% CI: -2.9, 6.3%) using Intake24, 1.3% (95% CI: -1.1, 3.8%) using mFR-TA, and 15.0% (95% CI: 11.6, 18.3%) using IA-24HR. The variances of estimated and true energy intakes were statistically significantly different for all methods (P < 0.01) except Intake24 (P = 0.1). Differential accuracy in nutrient estimation was present among the methods.

CONCLUSIONS

Under controlled conditions, Intake24, ASA24, and mFR-TA estimated average energy and nutrient intakes with reasonable validity, but intake distributions were estimated accurately by Intake24 only (energy and protein). This study may inform considerations regarding instruments of choice in future population surveillance. This trial was registered at Australian New Zealand Clinical Trials Registry as ACTRN12621000209897.

Effect of Intermittent vs. Continuous Energy Restriction on Visceral Fat: Protocol for The Healthy Diet and Lifestyle Study 2 (HDLS2)

Obesity in the United States and Western countries represents a major health challenge associated with an increased risk of metabolic diseases, including cardiovascular disease, hypertension, diabetes, and certain cancers. Our past work revealed a more pronounced obesity-cancer link in certain ethnic groups, motivating us to develop a tailored dietary intervention called the Healthy Diet and Lifestyle 2 (HDLS2). The study protocol is described herein for this randomized six-month trial examining the effects of intermittent energy restriction (5:2 Diet) plus the Mediterranean dietary pattern (IER + MED) on visceral adipose tissue (VAT), liver fat, and metabolic biomarkers, compared to a standard MED with daily energy restriction (DER + MED), in a diverse participant group. Using MRI and DXA scans for body composition analysis, as well as metabolic profiling, this research aims to contribute to nutritional guidelines and strategies for visceral obesity reduction. The potential benefits of IER + MED, particularly regarding VAT reduction and metabolic health improvement, could be pivotal in mitigating the obesity epidemic and its metabolic sequelae. The ongoing study will provide essential insights into the efficacy of these energy restriction approaches across varied racial/ethnic backgrounds, addressing an urgent need in nutrition and metabolic health research. Registered Trial, National Institutes of Health, ClinicalTrials.gov (NCT05132686).

Tailored Prompting to Improve Adherence to Image-Based Dietary Assessment: Mixed Methods Study

Background

Accurately assessing an individual's diet is vital in the management of personal nutrition and in the study of the effect of diet on health. Despite its importance, the tools available for dietary assessment remain either too imprecise, expensive, or burdensome for clinical or research use. Image-based methods offer a potential new tool to improve the reliability and accessibility of dietary assessment. Though promising, image-based methods are sensitive to adherence, as images cannot be captured from meals that have already been consumed. Adherence to image-based methods may be improved with appropriately timed prompting via text message.

Objective

This study aimed to quantitatively examine the effect of prompt timing on adherence to an image-based dietary record and qualitatively explore the participant experience of dietary assessment in order to inform the design of a novel image-based dietary assessment tool.

Methods

This study used a randomized crossover design to examine the intraindividual effect of 3 prompt settings on the number of images captured in an image-based dietary record. The prompt settings were control, where no prompts were sent; standard, where prompts were sent at 7:15 AM, 11:15 AM, and 5:15 PM for every participant; and tailored, where prompt timing was tailored to habitual meal times for each participant. Participants completed a text-based dietary record at baseline to determine the timing of tailored prompts. Participants were randomized to 1 of 6 study sequences, each with a unique order of the 3 prompt settings, with each 3-day image-based dietary record separated by a washout period of at least 7 days. The qualitative component comprised semistructured interviews and questionnaires exploring the experience of dietary assessment.

Results

A total of 37 people were recruited, and 30 participants (11 male, 19 female; mean age 30, SD 10.8 years), completed all image-based dietary records. The image rate increased by 0.83 images per day in the standard setting compared to control (P=.23) and increased by 1.78 images per day in the tailored setting compared to control (P≤.001). We found that 13/21 (62%) of participants preferred to use the image-based dietary record versus the text-based dietary record but reported method-specific challenges with each method, particularly the inability to record via an image after a meal had been consumed.

Conclusions

Tailored prompting improves adherence to image-based dietary assessment. Future image-based dietary assessment tools should use tailored prompting and offer both image-based and written input options to improve record completeness.

Perspective: Promoting Healthy Aging through Nutrition: A Research Centers Collaborative Network Workshop Report

Within 20 y, the number of adults in the United States over the age of 65 y is expected to more than double and the number over age 85 y is expected to more than triple. The risk for most chronic diseases and disabilities increases with age, so this demographic shift carries significant implications for the individual, health care providers, and population health. Strategies that delay or prevent the onset of age-related diseases are becoming increasingly important. Although considerable progress has been made in understanding the contribution of nutrition to healthy aging, it has become increasingly apparent that much remains to be learned, especially because the aging process is highly variable. Most federal nutrition programs and nutrition research studies define all adults over age 65 y as "older" and do not account for physiological and metabolic changes that occur throughout older adulthood that influence nutritional needs. Moreover, the older adult population is becoming more racially and ethnically diverse, so cultural preferences and other social determinants of health need to be considered. The Research Centers Collaborative Network sponsored a 1.5-d multidisciplinary workshop that included sessions on dietary patterns in health and disease, timing and targeting interventions, and health disparities and the social context of diet and food choice. The agenda and presentations can be found at https://www.rccn-aging.org/nutrition-2023-rccn-workshop. Here we summarize the workshop's themes and discussions and highlight research gaps that if filled will considerably advance our understanding of the role of nutrition in healthy aging.

Dietary Intake Assessment Using a Novel, Generic Meal-Based Recall and a 24-Hour Recall: Comparison Study

BACKGROUND

Dietary intake assessment is an integral part of addressing suboptimal dietary intakes. Existing food-based methods are time-consuming and burdensome for users to report the individual foods consumed at each meal. However, ease of use is the most important feature for individuals choosing a nutrition or diet app. Intakes of whole meals can be reported in a manner that is less burdensome than reporting individual foods. No study has developed a method of dietary intake assessment where individuals report their dietary intakes as whole meals rather than individual foods.

OBJECTIVE

This study aims to develop a novel, meal-based method of dietary intake assessment and test its ability to estimate nutrient intakes compared with that of a web-based, 24-hour recall (24HR).

METHODS

Participants completed a web-based, generic meal-based recall. This involved, for each meal type (breakfast, light meal, main meal, snack, and beverage), choosing from a selection of meal images those that most represented their intakes during the previous day. Meal images were based on generic meals from a previous study that were representative of the actual meal intakes in Ireland. Participants also completed a web-based 24HR. Both methods were completed on the same day, 3 hours apart. In a crossover design, participants were randomized in terms of which method they completed first. Then, 2 weeks after the first dietary assessments, participants repeated the process in the reverse order. Estimates of mean daily nutrient intakes and the categorization of individuals according to nutrient-based guidelines (eg, low, adequate, and high) were compared between the 2 methods. P values of less than .05 were considered statistically significant.

RESULTS

In total, 161 participants completed the study. For the 23 nutrient variables compared, the median percentage difference between the 2 methods was 7.6% (IQR 2.6%-13.2%), with P values ranging from <.001 to .97, and out of 23 variables, effect sizes for the differences were small for 19 (83%) variables, moderate for 2 (9%) variables, and large for 2 (9%) variables. Correlation coefficients were statistically significant (P<.05) for 18 (78%) of the 23 variables. Statistically significant correlations ranged from 0.16 to 0.45, with median correlation of 0.32 (IQR 0.25-0.40). When participants were classified according to nutrient-based guidelines, the proportion of individuals who were classified into the same category ranged from 52.8% (85/161) to 84.5% (136/161).

CONCLUSIONS

A generic meal-based method of dietary intake assessment provides estimates of nutrient intake comparable with those provided by a web-based 24HR but with varying levels of agreement among nutrients. Further studies are required to refine and improve the generic recall across a range of nutrients. Future studies will consider user experience including the potential feasibility of incorporating image recognition of whole meals into the generic recall.

Integrated image and sensor-based food intake detection in free-living

The first step in any dietary monitoring system is the automatic detection of eating episodes. To detect eating episodes, either sensor data or images can be used, and either method can result in false-positive detection. This study aims to reduce the number of false positives in the detection of eating episodes by a wearable sensor, Automatic Ingestion Monitor v2 (AIM-2). Thirty participants wore the AIM-2 for two days each (pseudo-free-living and free-living). The eating episodes were detected by three methods: (1) recognition of solid foods and beverages in images captured by AIM-2; (2) recognition of chewing from the AIM-2 accelerometer sensor; and (3) hierarchical classification to combine confidence scores from image and accelerometer classifiers. The integration of image- and sensor-based methods achieved 94.59% sensitivity, 70.47% precision, and 80.77% F1-score in the free-living environment, which is significantly better than either of the original methods (8% higher sensitivity). The proposed method successfully reduces the number of false positives in the detection of eating episodes.

The Validity and Feasibility of Utilizing the Photo-Assisted Dietary Intake Assessment among College Students and Elderly Individuals in China

Dietary assessments hold significant importance within the field of public health. However, the current methods employed for dietary assessments face certain limitations and challenges that necessitate improvement. The aim of our study was to develop a reliable and practical dietary assessment tool known as photo-assisted dietary intake assessment (PAD). In order to evaluate its validity, we conducted an analysis on a sample of 71 college students' dinners at a buffet in a canteen. We compared estimates of food weights obtained through the 24-h recall (24 HR) or PAD method with those obtained through the weighing method; we also evaluated the feasibility of PAD for recording dinner intakes among a sample of college students (n = 76) and elderly individuals (n = 121). In addition, we successfully identified the dietary factors that have a significant impact on the bias observed in weight estimation. The findings of the study indicated that the PAD method exhibited a higher level of consistency with the weighing method compared to the 24 HR method. The discrepancy in D% values between cereals (14.28% vs. 40.59%, P < 0.05), vegetables (17.67% vs. 44.44%, P < 0.05), and meats (14.29% vs. 33.33%, P < 0.05) was clearly apparent. Moreover, a significant proportion of the food mass value acquired through the PAD method fell within the limits of agreement (LOAs), in closer proximity to the central horizontal line. Furthermore, vegetables, cereals, eggs, and meats, for which the primary importance lies in accuracy, exhibited a considerably higher bias with the 24 HR method compared to the PAD method (P < 0.05), implying that the PAD method has the potential to mitigate the quality bias associated with these food items in the 24 HR method. Additionally, the PAD method was well received and easily implemented by the college students and elderly individuals. In conclusion, the PAD method demonstrates a considerable level of accuracy and feasibility as a dietary assessment method that can be effectively employed across diverse populations.

Development of a dental diet-tracking mobile app for improved caries-related dietary behaviours: Key features and pilot evaluation of quality

Objective

Diet significantly contributes to dental decay (caries) yet monitoring and modifying patients' diets is a challenge for many dental practitioners. While many oral health and diet-tracking mHealth apps are available, few focus on the dietary risk factors for caries. This study aims to present the development and key features of a dental-specific mobile app for diet monitoring and dietary behaviour change to prevent caries, and pilot data from initial user evaluation.

Methods

A mobile app incorporating a novel photo recognition algorithm and a localised database of 208,718 images for food item identification was developed. The design and development process were iterative and incorporated several behaviour change techniques commonly used in mHealth. Pilot evaluation of app quality was assessed using the end-user version of the Mobile Application Rating Scale (uMARS).

Results

User feedback from the beta-testing of the prototype app spurred the improvement of the photo recognition algorithm and addition of more user-centric features. Other key features of the final app include real-time prompts to drive actionable behaviour change, goal setting, comprehensive oral health education modules, and visual metrics for caries-related dietary factors (sugar intake, meal frequency, etc.). The final app scored an overall mean (standard deviation) of 3.6 (0.5) out of 5 on the uMARS scale.

Conclusion

We developed a novel diet-tracking mobile app tailored for oral health, addressing a gap in the mHealth landscape. Pilot user evaluations indicated good app quality, suggesting its potential as a useful clinical tool for dentists and empowering patients for self-monitoring and behavioural management.

Mindful Eco-Wellness: Steps Toward Personal and Planetary Health

Rising greenhouse gas levels heat the earth's surface and alter climate patterns, posing unprecedented threats to planetary ecology and human health. At the same time, obesity, diabetes, and cardiovascular disease have reached epidemic proportions across the globe, caused in part by decreases in physical activity and by over-consumption of carbon-intensive foods. Thus, interventions that support active transportation (walking or cycling rather than driving) and healthier food choices (eating plant-based rather than meat-based diets) would yield health and sustainability "co-benefits." Emerging research suggests that mindfulness-based practices might be effective means toward these ends. At the University of Wisconsin-Madison, we have developed a mindfulness-based group program, Mindful Eco-Wellness: Steps Toward Healthier Living. Loosely based on the Mindfulness-Based Stress Reduction course, our curriculum teaches mindfulness practices in tandem with sustainability principles, following weekly themes of Air, Water, Food, Energy, Transportation, Consumption, Nature Experience, and Ethics. For example, the "Air" class offers participants practice in guided breath meditations while they learn about the benefits of clean air. The theme of "Food" is presented through mindful eating, accompanied by educational videos highlighting the consequences of food production and consumption. "Transportation" includes walking/movement meditations and highlights the health benefits of physical activity and detriments of fossil-fueled transportation. Pedagogical lessons on energy, ecological sustainability, and the ethics of planetary health are intertwined with mindful nature experience and metta (loving-kindness) meditation. Curricular materials, including teaching videos, are freely available online. Pilot testing in community settings (n = 30) and in group medical visits (n = 34) has demonstrated feasibility; pilot data suggests potential effectiveness. Rigorous evaluation and testing are needed.

Dietary assessment methods for measurement of oral intake in acute care and critically ill hospitalised patients: a scoping review

Quantification of oral intake within the hospital setting is required to guide nutrition care. Multiple dietary assessment methods are available, yet details regarding their application in the acute care setting are scarce. This scoping review, conducted in accordance with JBI methodology, describes dietary assessment methods used to measure oral intake in acute and critical care hospital patients. The search was run across four databases to identify primary research conducted in adult acute or critical care settings from 1st of January 2000-15th March 2023 which quantified oral diet with any dietary assessment method. In total, 155 articles were included, predominantly from the acute care setting (n = 153, 99%). Studies were mainly single-centre (n = 138, 88%) and of observational design (n = 135, 87%). Estimated plate waste (n = 59, 38%) and food records (n = 43, 28%) were the most frequent assessment methods with energy and protein the main nutrients quantified (n = 81, 52%). Validation was completed in 23 (15%) studies, with the majority of these using a reference method reliant on estimation (n = 17, 74%). A quarter of studies (n = 39) quantified completion (either as complete versus incomplete or degree of completeness) and four studies (2.5%) explored factors influencing completion. Findings indicate a lack of high-quality evidence to guide selection and application of existing dietary assessment methods to quantify oral intake with a particular absence of evidence in the critical care setting. Further validation of existing tools and identification of factors influencing completion is needed to guide the optimal approach to quantification of oral intake in both research and clinical contexts.

AI-based digital image dietary assessment methods compared to humans and ground truth: a systematic review

OBJECTIVE

Human error estimating food intake is a major source of bias in nutrition research. Artificial intelligence (AI) methods may reduce bias, but the overall accuracy of AI estimates is unknown. This study was a systematic review of peer-reviewed journal articles comparing fully automated AI-based (e.g. deep learning) methods of dietary assessment from digital images to human assessors and ground truth (e.g. doubly labelled water).

MATERIALS AND METHODS

Literature was searched through May 2023 in four electronic databases plus reference mining. Eligible articles reported AI estimated volume, energy, or nutrients. Independent investigators screened articles and extracted data. Potential sources of bias were documented in absence of an applicable risk of bias assessment tool.

RESULTS

Database and hand searches identified 14,059 unique publications; fifty-two papers (studies) published from 2010 to 2023 were retained. For food detection and classification, 79% of papers used a convolutional neural network. Common ground truth sources were calculation using nutrient tables (51%) and weighed food (27%). Included papers varied widely in food image databases and results reported, so meta-analytic synthesis could not be conducted. Relative errors were extracted or calculated from 69% of papers. Average overall relative errors (AI vs. ground truth) ranged from 0.10% to 38.3% for calories and 0.09% to 33% for volume, suggesting similar performance. Ranges of relative error were lower when images had single/simple foods.

CONCLUSIONS

Relative errors for volume and calorie estimations suggest that AI methods align with - and have the potential to exceed - accuracy of human estimations. However, variability in food image databases and results reported prevented meta-analytic synthesis. The field can advance by testing AI architectures on a limited number of large-scale food image and nutrition databases that the field determines to be adequate for training and testing and by reporting accuracy of at least absolute and relative error for volume or calorie estimations.

Mitigating errors in mobile-based dietary assessments: Effects of a data modification process on the validity of an image-assisted food and nutrition app

OBJECTIVE

Mobile nutrition applications (apps) provide a simple way for individuals to record their diet, but the validity and inherent errors need to be carefully evaluated. The aim of this study was to assess the validity and clarify the sources of measurement errors of image-assisted mobile nutrition apps.

METHODS

This was a cross-sectional study with 98 students recruited from School of Nutrition and Health Sciences, Taipei Medical University. A 3-d nutrient intake record by Formosa Food and Nutrient Recording App (FoodApp) was compared with a 24-h dietary recall (24-HDR). A two-stage data modification process, manual data cleaning, and reanalyzing of prepackaged foods were employed to address inherent errors. Nutrient intake levels obtained by the two methods were compared with the recommended daily intake (DRI), Taiwan. Paired t test, Spearman's correlation coefficients, and Bland-Altman plots were used to assess agreement between the FoodApp and 24-HDR.

RESULTS

Manual data cleaning identified 166 food coding errors (12%; stage 1), and 426 food codes with missing micronutrients (32%) were reanalyzed (stage 2). Positive linear trends were observed for total energy and micronutrient intake (all Ptrend < 0.05) after the two stages of data modification, but not for dietary fat, carbohydrates, or vitamin D. There were no statistical differences in mean energy and macronutrient intake between the FoodApp and 24-HDR, and this agreement was confirmed by Bland-Altman plots. Spearman's correlation analyses showed strong to moderate correlations (r = 0.834 ∼ 0.386) between the two methods. Participants' nutrient intake tended to be lower than the DRI, but no differences in proportions of adequacy/inadequacy for DRI values were observed between the two methods.

CONCLUSIONS

Mitigating errors significantly improved the accuracy of the Formosa FoodApp, indicating its validity and reliability as a self-reporting mobile-based dietary assessment tool. Dietitians and health professionals should be mindful of potential errors associated with self-reporting nutrition apps, and manual data cleaning is vital to obtain reliable nutrient intake data.

Surveying Nutrient Assessment with Photographs of Meals (SNAPMe): A Benchmark Dataset of Food Photos for Dietary Assessment

Photo-based dietary assessment is becoming more feasible as artificial intelligence methods improve. However, advancement of these methods for dietary assessment in research settings has been hindered by the lack of an appropriate dataset against which to benchmark algorithm performance. We conducted the Surveying Nutrient Assessment with Photographs of Meals (SNAPMe) study (ClinicalTrials ID: NCT05008653) to pair meal photographs with traditional food records. Participants were recruited nationally, and 110 enrollment meetings were completed via web-based video conferencing. Participants uploaded and annotated their meal photos using a mobile phone app called Bitesnap and completed food records using the Automated Self-Administered 24-h Dietary Assessment Tool (ASA24®) version 2020. Participants included photos before and after eating non-packaged and multi-serving packaged meals, as well as photos of the front and ingredient labels for single-serving packaged foods. The SNAPMe Database (DB) contains 3311 unique food photos linked with 275 ASA24 food records from 95 participants who photographed all foods consumed and recorded food records in parallel for up to 3 study days each. The use of the SNAPMe DB to evaluate ingredient prediction demonstrated that the publicly available algorithms FB Inverse Cooking and Im2Recipe performed poorly, especially for single-ingredient foods and beverages. Correlations between nutrient estimates common to the Bitesnap and ASA24 dietary assessment tools indicated a range in predictive capacity across nutrients (cholesterol, adjusted R2 = 0.85, p < 0.0001; food folate, adjusted R2 = 0.21, p < 0.05). SNAPMe DB is a publicly available benchmark for photo-based dietary assessment in nutrition research. Its demonstrated utility suggested areas of needed improvement, especially the prediction of single-ingredient foods and beverages.

Sleep Variability, Eating Timing Variability, and Carotid Intima-Media Thickness in Early Adulthood

Background Day-to-day variability in sleep patterns and eating timing may disrupt circadian rhythms and has been linked with various adverse cardiometabolic outcomes. However, the extent to which variability in sleep patterns and eating timing relate to atherosclerotic development in subclinical stages remains unclear. Methods and Results Generally healthy adults (N=62, 29.3±7.3 years, 66% female) completed 14 days of sleep and dietary assessments via wrist accelerometry and photo-assisted diet records, respectively. Variability in sleep duration, sleep onset, eating onset (time of first caloric consumption), eating offset (time of last caloric consumption), and caloric midpoint (time at which 50% of total daily calories are consumed) were operationalized as the SD across 14 days for each variable. Separate regression models evaluated the cross-sectional associations between sleep and eating variability metrics with end-diastolic carotid intima-media thickness (CIMT) measured via ultrasonography. Models adjusted for age, sex, systolic blood pressure, sleep duration, and total energy intake. Each 60-minute increase in sleep duration SD and sleep onset SD were associated with a 0.049±0.016 mm (P=0.003) and 0.048±0.017 mm (P=0.007) greater CIMT, respectively. Variability in eating onset and offset were not associated with CIMT; however, each 60-minute increase in caloric midpoint SD was associated with a 0.033±0.015 mm greater CIMT (P=0.029). Exploratory post hoc analyses suggested that sleep duration SD and sleep onset SD were stronger correlates of CIMT than caloric midpoint SD. Conclusions Variability in sleep patterns and eating timing are positively associated with clinically relevant increases in CIMT, a biomarker of subclinical atherosclerosis, in early adulthood.

Estimating Amount of Food in a Circular Dining Bowl from a Single Image

Unhealthy diet is a top risk factor causing obesity and numerous chronic diseases. To help the public adopt healthy diet, nutrition scientists need user-friendly tools to conduct Dietary Assessment (DA). In recent years, new DA tools have been developed using a smartphone or a wearable device which acquires images during a meal. These images are then processed to estimate calories and nutrients of the consumed food. Although considerable progress has been made, 2D food images lack scale reference and 3D volumetric information. In addition, food must be sufficiently observable from the image. This basic condition can be met when the food is stand-alone (no food container is used) or it is contained in a shallow plate. However, the condition cannot be met easily when a bowl is used. The food is often occluded by the bowl edge, and the shape of the bowl may not be fully determined from the image. However, bowls are the most utilized food containers by billions of people in many parts of the world, especially in Asia and Africa. In this work, we propose to premeasure plates and bowls using a marked adhesive strip before a dietary study starts. This simple procedure eliminates the use of a scale reference throughout the DA study. In addition, we use mathematical models and image processing to reconstruct the bowl in 3D. Our key idea is to estimate how full the bowl is rather than how much food is (in either volume or weight) in the bowl. This idea reduces the effect of occlusion. The experimental data have shown satisfactory results of our methods which enable accurate DA studies using both plates and bowls with reduced burden on research participants.

Assessing daily energy intake in adult women: validity of a food-recognition mobile application compared to doubly labelled water

Accurate dietary assessment is crucial for nutrition and health research. Traditional methods, such as food records, food frequency questionnaires, and 24-hour dietary recalls (24HR), have limitations, such as the need for trained interviewers, time-consuming procedures, and inaccuracies in estimations. Novel technologies, such as image-based dietary assessment apps, have been developed to overcome these limitations. SNAQ is a novel image-based food-recognition app which, based on computer vision, assesses food type and volume, and provides nutritional information about dietary intake. This cross-sectional observational study aimed to investigate the validity of SNAQ as a dietary assessment tool for measuring energy and macronutrient intake in adult women with normal body weight (n = 30), compared to doubly labeled water (DLW), a reference method for total daily energy expenditure (TDEE). Energy intake was also estimated using a one-day 24HR for direct comparison. Bland-Altman plots, paired difference tests, and Pearson's correlation coefficient were used to assess agreement and relationships between the methods. SNAQ showed a slightly higher agreement (bias = -329.6 kcal/day) with DLW for total daily energy intake (TDEI) compared to 24HR (bias = -543.0 kcal/day). While both SNAQ and 24HR tended to underestimate TDEI, only 24HR significantly differed from DLW in this regard (p < 0.001). There was no significant relationship between estimated TDEI and TDEE using SNAQ (R2 = 27%, p = 0.50) or 24HR (R2 = 34%, p = 0.20) and there were no significant differences in energy and macronutrient intake estimates between SNAQ and 24HR (Δ = 213.4 kcal/day). In conclusion, these results indicate that SNAQ provides a closer representation of energy intake in adult women with normal body weight than 24HR when compared to DLW, but no relationship was found between the energy estimates of DLW and of the two dietary assessment tools. Further research is needed to determine the clinical relevance and support the implementation of SNAQ in research and clinical settings. Clinical trial registration: This study is registered on ClinicalTrials.gov with the unique identifier NCT04600596 (https://clinicaltrials.gov/ct2/show/NCT04600596).

Evaluation of Acceptability, Functionality, and Validity of a Passive Image-Based Dietary Intake Assessment Method in Adults and Children of Ghanaian and Kenyan Origin Living in London, UK

BACKGROUND

Accurate estimation of dietary intake is challenging. However, whilst some progress has been made in high-income countries, low- and middle-income countries (LMICs) remain behind, contributing to critical nutritional data gaps. This study aimed to validate an objective, passive image-based dietary intake assessment method against weighed food records in London, UK, for onward deployment to LMICs.

METHODS

Wearable camera devices were used to capture food intake on eating occasions in 18 adults and 17 children of Ghanaian and Kenyan origin living in London. Participants were provided pre-weighed meals of Ghanaian and Kenyan cuisine and camera devices to automatically capture images of the eating occasions. Food images were assessed for portion size, energy, nutrient intake, and the relative validity of the method compared to the weighed food records.

RESULTS

The Pearson and Intraclass correlation coefficients of estimates of intakes of food, energy, and 19 nutrients ranged from 0.60 to 0.95 and 0.67 to 0.90, respectively. Bland-Altman analysis showed good agreement between the image-based method and the weighed food record. Under-estimation of dietary intake by the image-based method ranged from 4 to 23%.

CONCLUSIONS

Passive food image capture and analysis provides an objective assessment of dietary intake comparable to weighed food records.

Contribution of Different Food Types to Vitamin A Intake in the Chinese Diet

Vitamin A is a fat-soluble micronutrient that is essential for human health. In this study, the daily vitamin A intake of Chinese residents was evaluated by investigating the vitamin A content of various foods. The results show that the dietary intake of vitamin A in common foods was 460.56 ugRAE/day, which is significantly lower than the recommended dietary reference intake of vitamin A (800 ugRAE/day for adult men and 700 ugRAE/day for adult women). Vegetables contributed the most to daily vitamin A dietary intake, accounting for 54.94% of vitamin A intake (253.03 ugRAE/day), followed by eggs, milk, aquatic products, meat, fruit, legumes, coarse cereals, and potatoes. Therefore, an increase in the vitamin A content of vegetables and the fortification of vegetable oils with vitamin A are effective ways to increase vitamin A intake to meet the recommended dietary guidelines in China. The assessment results support the design of fortified foods.

Perspective: Challenges and Future Directions in Clinical Research with Nuts and Berries

Consumption of nuts and berries are considered part of a healthy eating pattern. Nuts and berries contain a complex nutrient profile consisting of essential vitamins and minerals, fiber, polyunsaturated fatty acids, and phenolics in quantities that improve physiological outcomes. The spectrum of health outcomes that may be impacted by the consumptions of nuts and berries includes cardiovascular, gut microbiome, and cognitive, among others. Recently, new insights regarding the bioactive compounds found in both nuts and berries have reinforced their role for use in precision nutrition efforts. However, challenges exist that can affect the generalizability of outcomes from clinical studies, including inconsistency in study designs, homogeneity of test populations, variability in test products and control foods, and assessing realistic portion sizes. Future research centered on precision nutrition and multi-omics technologies will yield new insights. These and other topics such as funding streams and perceived risk-of-bias were explored at an international nutrition conference focused on the role of nuts and berries in clinical nutrition. Successes, challenges, and future directions with these foods are presented here.

Guidelines for microbiome studies in renal physiology

Gut microbiome research has increased dramatically in the last decade, including in renal health and disease. The field is moving from experiments showing mere association to causation using both forward and reverse microbiome approaches, leveraging tools such as germ-free animals, treatment with antibiotics, and fecal microbiota transplantations. However, we are still seeing a gap between discovery and translation that needs to be addressed, so that patients can benefit from microbiome-based therapies. In this guideline paper, we discuss the key considerations that affect the gut microbiome of animals and clinical studies assessing renal function, many of which are often overlooked, resulting in false-positive results. For animal studies, these include suppliers, acclimatization, baseline microbiota and its normalization, littermates and cohort/cage effects, diet, sex differences, age, circadian differences, antibiotics and sweeteners, and models used. Clinical studies have some unique considerations, which include sampling, gut transit time, dietary records, medication, and renal phenotypes. We provide best-practice guidance on sampling, storage, DNA extraction, and methods for microbial DNA sequencing (both 16S rRNA and shotgun metagenome). Finally, we discuss follow-up analyses, including tools available, metrics, and their interpretation, and the key challenges ahead in the microbiome field. By standardizing study designs, methods, and reporting, we will accelerate the findings from discovery to translation and result in new microbiome-based therapies that may improve renal health.

I2N: image to nutrients, a sensor guided semi-automated tool for annotation of images for nutrition analysis of eating episodes

Introduction

Dietary assessment is important for understanding nutritional status. Traditional methods of monitoring food intake through self-report such as diet diaries, 24-hour dietary recall, and food frequency questionnaires may be subject to errors and can be time-consuming for the user.

Methods

This paper presents a semi-automatic dietary assessment tool we developed - a desktop application called Image to Nutrients (I2N) - to process sensor-detected eating events and images captured during these eating events by a wearable sensor. I2N has the capacity to offer multiple food and nutrient databases (e.g., USDA-SR, FNDDS, USDA Global Branded Food Products Database) for annotating eating episodes and food items. I2N estimates energy intake, nutritional content, and the amount consumed. The components of I2N are three-fold: 1) sensor-guided image review, 2) annotation of food images for nutritional analysis, and 3) access to multiple food databases. Two studies were used to evaluate the feasibility and usefulness of I2N: 1) a US-based study with 30 participants and a total of 60 days of data and 2) a Ghana-based study with 41 participants and a total of 41 days of data).

Results

In both studies, a total of 314 eating episodes were annotated using at least three food databases. Using I2N's sensor-guided image review, the number of images that needed to be reviewed was reduced by 93% and 85% for the two studies, respectively, compared to reviewing all the images.

Discussion

I2N is a unique tool that allows for simultaneous viewing of food images, sensor-guided image review, and access to multiple databases in one tool, making nutritional analysis of food images efficient. The tool is flexible, allowing for nutritional analysis of images if sensor signals aren't available.

Integration of USDA Food Classification System and Food Composition Database for Image-Based Dietary Assessment among Individuals Using Insulin

New imaging technologies to identify food can reduce the reporting burden of participants but heavily rely on the quality of the food image databases to which they are linked to accurately identify food images. The objective of this study was to develop methods to create a food image database based on the most commonly consumed U.S. foods and those contributing the most to energy. The objective included using a systematic classification structure for foods based on the standardized United States Department of Agriculture (USDA) What We Eat in America (WWEIA) food classification system that can ultimately be used to link food images to a nutrition composition database, the USDA Food and Nutrient Database for Dietary Studies (FNDDS). The food image database was built using images mined from the web that were fitted with bounding boxes, identified, annotated, and then organized according to classifications aligning with USDA WWEIA. The images were classified by food category and subcategory and then assigned a corresponding USDA food code within the USDA's FNDDS in order to systematically organize the food images and facilitate a linkage to nutrient composition. The resulting food image database can be used in food identification and dietary assessment.

Long-Tailed Food Classification

Food classification serves as the basic step of image-based dietary assessment to predict the types of foods in each input image. However, foods in real-world scenarios are typically long-tail distributed, where a small number of food types are consumed more frequently than others, which causes a severe class imbalance issue and hinders the overall performance. In addition, none of the existing long-tailed classification methods focus on food data, which can be more challenging due to the inter-class similarity and intra-class diversity between food images. In this work, two new benchmark datasets for long-tailed food classification are introduced, including Food101-LT and VFN-LT, where the number of samples in VFN-LT exhibits real-world long-tailed food distribution. Then, a novel two-phase framework is proposed to address the problem of class imbalance by (1) undersampling the head classes to remove redundant samples along with maintaining the learned information through knowledge distillation and (2) oversampling the tail classes by performing visually aware data augmentation. By comparing our method with existing state-of-the-art long-tailed classification methods, we show the effectiveness of the proposed framework, which obtains the best performance on both Food101-LT and VFN-LT datasets. The results demonstrate the potential to apply the proposed method to related real-life applications.

Feasibility and acceptability of a culturally-adapted Women's Wellness After Cancer Programme for Chinese women treated for gynaecological cancer: A pilot randomised controlled trial

Objective

To assess the feasibility and acceptability of a culturally-adapted Women's Wellness After Cancer Programme (WWACPHK) for improving health-related quality of life, anxiety and depressive symptoms and enhancing self-efficacy in engaging in healthy lifestyles among Chinese women treated for gynaecological cancer.

Methods

This pilot randomised controlled trial was conducted from May to December 2018. Twenty-six women aged 18 or above who had completed treatment for gynaecological cancer were recruited from a gynaecology outpatient clinic of a public hospital in Hong Kong. They were randomised into intervention (n = 15) or control (n = 11) groups. All data collectors were blinded to the group allocation. Intervention participants were given access to the WWACPHK website and an online discussion forum facilitated by a trained research nurse for 12 weeks, while control participants received standard care. Trial feasibility was assessed by recruitment, consent, and retention rates and website use. Acceptability was explored through semi-structured interviews. Additionally, we trialed the data collection procedure and collected preliminary data on health-related quality of life, anxiety and depressive symptoms, dietary and exercise self-efficacy.

Results

Of the 26 participants (Median age = 53.5 years) randomised, three participants dropped out of the study. Recruitment, consent and retention of participants and website use were satisfactory. No posting was made on the discussion forum. The intervention participants (n = 13) exhibited significantly greater improvement than the controls (n = 10) in perceived self-efficacy in adhering to an exercise routine at post-intervention (Cohen's d effect size(d) = 1.06, 95% confidence interval (CI): 0.18, 1.92) and 12-weeks after completion (d = 1.24, 95% CI: 0.32, 2.13). All participants were satisfied with the intervention.

Conclusions

The WWACPHK is feasible and acceptable to Chinese women treated for gynaecological cancer and may improve their exercise self-efficacy. A larger-scale study is required to confirm its effects. Trial registrationhttps://www.isrctn.com identifier: ISRCTN12149499.

Relative Validity and Reliability of the Remind App as an Image-Based Method to Assess Dietary Intake and Meal Timing in Young Adults

Image-based dietary records have been validated as tools to evaluate dietary intake. However, to determine meal timing, previous studies have relied primarily on image-based smartphone applications without validation. Noteworthy, the validation process is necessary to determine how accurately a test method measures meal timing compared with a reference method over the same time period. Thus, we aimed to assess the relative validity and reliability of the Remind^® app as an image-based method to assess dietary intake and meal timing. For this purpose, 71 young adults (aged 20-33 years, 81.7% women) were recruited for a 3-day cross-sectional study, where they completed a 3-day image-based record using the Remind app (test method) and a 3-day handwritten food record (reference method). The relative validity of the test method versus the reference method was assessed using multiple tests including Bland-Altman, % difference, paired t-test/Wilcoxon signed-rank test, Pearson/Spearman correlation coefficients, and cross-classification. We also evaluated the reliability of the test method using an intra-class correlation (ICC) coefficient. The results showed that, compared to the reference method, the relative validity of the test method was good for assessing energy and macronutrient intake, as well as meal timing. Meanwhile, the relative validity of the test method to assess micronutrient intake was poor (p < 0.05) for some micronutrients (iron, phosphorus, potassium, zinc, vitamins B1, B2, B3, B6, C, and E, and folates) and some food groups (cereals and grains, legumes, tubers, oils, and fats). Regarding the reliability of an image-based method to assess dietary intake and meal timing, results ranged from moderate to excellent (ICC 95% confidence interval [95% CI]: 0.50-1.00) for all nutrients, food groups (except oils and fats, which had low to moderate reliability), and meal timings. Thus, the results obtained in this study provide evidence of the relative validity and reliability of image-based methods to assess dietary intake (energy, macronutrients, and most food groups) and meal timing. These results open up a new framework for chrononutrition, as these methods improve the quality of the data collected and also reduce the burden on users to accurately estimate portion size and the timing of meals.

A Diet Profiling Algorithm (DPA) to Rank Diet Quality Suitable to Implement in Digital Tools—A Test Study in a Cohort of Lactating Women

Although nutrient profiling systems can empower consumers towards healthier food choices, there is still a need to assess diet quality to obtain an overall perspective. The purpose of this study was to develop a diet profiling algorithm (DPA) to evaluate nutritional diet quality, which gives a final score from 1 to 3 with an associated color (green-yellow-orange). It ranks the total carbohydrate/total fiber ratio, and energy from saturated fats and sodium as potentially negative inputs, while fiber and protein are assumed as positive items. Then, the total fat/total carbohydrate ratio is calculated to evaluate the macronutrient distribution, as well as a food group analysis. To test the DPA performance, diets of a lactating women cohort were analyzed, and a correlation analysis between DPA and breast milk leptin levels was performed. Diets classified as low quality showed a higher intake of negative inputs, along with higher energy and fat intakes. This was reflected in body mass index (BMI) and food groups, indicating that women with the worst scores tended to choose tastier and less satiating foods. In conclusion, the DPA was developed and tested in a sample population. This tool can be easily implemented in digital nutrition platforms, contributing to real-time dietary follow-up of patients and progress monitoring, leading to further dietary adjustment.

Fecal Microbiota Composition as a Metagenomic Biomarker of Dietary Intake

Gut microbiota encompasses the set of microorganisms that colonize the gastrointestinal tract with mutual relationships that are key for host homeostasis. Increasing evidence supports cross intercommunication between the intestinal microbiome and the eubiosis-dysbiosis binomial, indicating a networking role of gut bacteria as potential metabolic health surrogate markers. The abundance and diversity of the fecal microbial community are already recognized to be associated with several disorders, such as obesity, cardiometabolic events, gastrointestinal alterations, and mental diseases, which suggests that intestinal microbes may be a valuable tool as causal or as consequence biomarkers. In this context, the fecal microbiota could also be used as an adequate and informative proxy of the nutritional composition of the food intake and about the adherence to dietary patterns, such as the Mediterranean or Western diets, by displaying specific fecal microbiome signatures. The aim of this review was to discuss the potential use of gut microbial composition as a putative biomarker of food intake and to screen the sensitivity value of fecal microbiota in the evaluation of dietary interventions as a reliable and precise alternative to subjective questionnaires.

Estimation of Plate and Bowl Dimensions for Food Portion Size Assessment in a Wearable Sensor System

Automatic food portion size estimation (FPSE) with minimal user burden is a challenging task. Most of the existing FPSE methods use fiducial markers and/or virtual models as dimensional references. An alternative approach is to estimate the dimensions of the eating containers prior to estimating the portion size. In this article, we propose a wearable sensor system (the automatic ingestion monitor integrated with a ranging sensor) and a related method for the estimation of dimensions of plates and bowls. The contributions of this study are: 1) the model eliminates the need for fiducial markers; 2) the camera system [automatic ingestion monitor version 2 (AIM-2)] is not restricted in terms of positioning relative to the food item; 3) our model accounts for radial lens distortion caused due to lens aberrations; 4) a ranging sensor directly gives the distance between the sensor and the eating surface; 5) the model is not restricted to circular plates; and 6) the proposed system implements a passive method that can be used for assessment of container dimensions with minimum user interaction. The error rates (mean ± std. dev) for dimension estimation were 2.01% ± 4.10% for plate widths/diameters, 2.75% ± 38.11% for bowl heights, and 4.58% ± 6.78% for bowl diameters.

Validation of the SmartPlate for detecting food weight and type

This study determined accuracy (comparing to criterion), inter-plate reliability (comparing measures between two plates), and intra-plate reliability (comparing successive measures on one plate) of the SmartPlate for food weight and type. Food weight validation included comparing SmartPlate weights to criterion [reference] scale weights (1,980 measures) and weights of 188 foods (2,256 measures). Food type validation included assessing SmartPlate accuracy for 188 foods. For weight, mean absolute percent errors for accuracy, inter-plate reliability, and intra-plate reliability were 6.2, 7.4, and 4.9%, respectively. For food type, foods were correctly identified/listed or searchable 67.0 or 98.9% of the time, respectively, with 76.0% inter-plate reliability and 86.3% intra-plate reliability. The SmartPlate had acceptable accuracy and reliability for assessing food weight and type and may be appealing for monitoring dietary surveillance or intervention. Due to high intra-plate reliability, the SmartPlate may be especially useful for one-on-one interventions and assessing change over time.

Validation of Diet ID™ in Predicting Nutrient Intake Compared to Dietary Recalls, Skin Carotenoid Scores, and Plasma Carotenoids in University Students

Background and Aim: Collecting accurate dietary information in the research setting is challenging due to the inherent biases, duration, and resource-intensive nature of traditional data collection methods. Diet ID™ is a novel, rapid assessment method that uses an image-based algorithm to identify dietary patterns and estimate nutrient intake. The purpose of this analysis was to explore the criterion validity between Diet ID™ and additional measures of dietary intake. Methods: This prospective cohort study (n = 42) collected dietary information using Diet ID™, the Nutrition Data System for Research (NDSR), plasma carotenoid concentrations, and the Veggie Meter^® to estimate carotenoid levels in the skin. Results: There were significant correlations between Diet ID™ and NDSR for diet quality, calories, carbohydrates, protein, fiber, and cholesterol. Vitamin A and carotenoid intake were significantly correlated, with the exception of α-carotene and lycopene. Significant correlations were observed for calcium, folate, iron, sodium, potassium, Vitamins B₂, B₃, B₆, C, and E. Skin carotenoid scores and plasma carotenoids were correlated with carotenoid intake from Diet ID™. Conclusions: Diet ID™ may be a useful tool in nutrition research as a less time-intensive and minimally burdensome dietary data collection method for both participants and researchers.

Development of a Web-Based Diabetes Prevention Program (DPP) for Chinese Americans: A Formative Evaluation Approach

Increasing evidence demonstrates that an online Diabetes Prevention Program (DPP) can delay the onset of type 2 diabetes. However, little has been done for Chinese Americans. This study, using Community-Based Participatory Research and Intervention Mapping approaches, describes a formative research process in the development of a culturally and linguistically tailored online DPP program among Chinese Americans with prediabetes living in New York City. Using a triangulation approach, data were collected to inform the development of an online DPP curriculum through (1) a literature review, (2) three focus groups (n = 24), and (3) a community advisory board meeting among 10 key informants knowledgeable in community needs, diabetes care, and lifestyle interventions. Participants indicated online DPPs would be very useful and easily accessible. However, key barriers including low computer skills/literacy and technology self-efficacy were identified. In addition, taking meal photos and tracking pedometer steps daily were found to be acceptable self-motoring tools for sustaining a healthy lifestyle. Furthermore, the integration of features such as text message reminders and the creation of social support groups into the online DPP curriculum was proposed to minimize attrition. This theory-based formative research to develop a culturally and linguistically appropriate web-based DPP curriculum was well-received by Chinese Americans and warrants testing in future intervention studies.

Mobile Computer Vision-Based Applications for Food Recognition and Volume and Calorific Estimation: A Systematic Review

The growing awareness of the influence of "what we eat" on lifestyle and health has led to an increase in the use of embedded food analysis and recognition systems. These solutions aim to effectively monitor daily food consumption, and therefore provide dietary recommendations to enable and support lifestyle changes. Mobile applications, due to their high accessibility, are ideal for real-life food recognition, volume estimation and calorific estimation. In this study, we conducted a systematic review based on articles that proposed mobile computer vision-based solutions for food recognition, volume estimation and calorific estimation. In addition, we assessed the extent to which these applications provide explanations to aid the users to understand the related classification and/or predictions. Our results show that 90.9% of applications do not distinguish between food and non-food. Similarly, only one study that proposed a mobile computer vision-based application for dietary intake attempted to provide explanations of features that contribute towards classification. Mobile computer vision-based applications are attracting a lot of interest in healthcare. They have the potential to assist in the management of chronic illnesses such as diabetes, ensuring that patients eat healthily and reducing complications associated with unhealthy food. However, to improve trust, mobile computer vision-based applications in healthcare should provide explanations of how they derive their classifications or volume and calorific estimations.

Applying Image-Based Food-Recognition Systems on Dietary Assessment: A Systematic Review

Dietary assessment can be crucial for the overall well-being of humans and, at least in some instances, for the prevention and management of chronic, life-threatening diseases. Recall and manual record-keeping methods for food-intake monitoring are available, but often inaccurate when applied for a long period of time. On the other hand, automatic record-keeping approaches that adopt mobile cameras and computer vision methods seem to simplify the process and can improve current human-centric diet-monitoring methods. Here we present an extended critical literature overview of image-based food-recognition systems (IBFRS) combining a camera of the user's mobile device with computer vision methods and publicly available food datasets (PAFDs). In brief, such systems consist of several phases, such as the segmentation of the food items on the plate, the classification of the food items in a specific food category, and the estimation phase of volume, calories, or nutrients of each food item. A total of 159 studies were screened in this systematic review of IBFRS. A detailed overview of the methods adopted in each of the 78 included studies of this systematic review of IBFRS is provided along with their performance on PAFDs. Studies that included IBFRS without presenting their performance in at least 1 of the above-mentioned phases were excluded. Among the included studies, 45 (58%) studies adopted deep learning methods and especially convolutional neural networks (CNNs) in at least 1 phase of the IBFRS with input PAFDs. Among the implemented techniques, CNNs outperform all other approaches on the PAFDs with a large volume of data, since the richness of these datasets provides adequate training resources for such algorithms. We also present evidence for the benefits of application of IBFRS in professional dietetic practice. Furthermore, challenges related to the IBFRS presented here are also thoroughly discussed along with future directions.

Effectiveness of dietary interventions delivered by digital health to adults with chronic conditions: Systematic review and meta-analysis

BACKGROUND

Digital health interventions may facilitate management of chronic conditions; however, no reviews have systematically assessed the effectiveness of dietary interventions delivered by digital health platforms for improving dietary intake and clinical outcomes for adults with diet-related chronic conditions.

METHODS

Databases CINAHL, CENTRAL, Embase and MEDLINE were searched from inception to April 2021 to identify controlled trials for dietary education delivered by digital health (mobile or electronic health) to adults with diet-related chronic conditions. Random effects analysis was performed for diet quality, food groups, nutrients and clinical outcomes. Screening, data extraction and quality checking were completed in duplicate.

RESULTS

Thirty-nine studies were included involving 7333 participants. Significant changes were found for Mediterranean diet adherence score (standardised mean difference [SMD] = 0.79; 95% confidence interval [CI] = 0.18 to 1.40), overall fruit and vegetable intake (mean difference [MD]: 0.63 serves/day; 95% CI = 0.27-0.98), fruit intake alone (MD = 0.58 serves/day; 95% CI = 0.39 to 0.77) and sodium intake (SMD = -0.22; 95% CI = -0.44 to -0.01). Improvements were also found for waist circumference [MD = -2.24 centimetres; 95% CI = -4.14 to -0.33], body weight (MD = -1.94 kg; 95% CI = -2.63 to -1.24) and haemoglobin A1c (MD = -0.17%; 95% CI = -0.29 to -0.04). Validity of digital assessment tools to measure dietary intake were not reported. The quality of evidence was considered to have low to moderate certainty.

CONCLUSIONS

Modest improvements in diet and clinical outcomes may result from intervention via digital health for those with diet-related chronic conditions. However, additional robust trials with better reporting of digital dietary assessment tools are needed to support implementation within clinical practice.

Characteristics of smartphone-based dietary assessment tools: a systematic review

Smartphones have become popular in assessing eating behaviour in real-life and real-time. This systematic review provides a comprehensive overview of smartphone-based dietary assessment tools, focusing on how dietary data is assessed and its completeness ensured. Seven databases from behavioural, social and computer science were searched in March 2020. All observational, experimental or intervention studies and study protocols using a smartphone-based assessment tool for dietary intake were included if they reported data collected by adults and were published in English. Out of 21,722 records initially screened, 117 publications using 129 tools were included. Five core assessment features were identified: photo-based assessment (48.8% of tools), assessed serving/ portion sizes (48.8%), free-text descriptions of food intake (42.6%), food databases (30.2%), and classification systems (27.9%). On average, a tool used two features. The majority of studies did not implement any features to improve completeness of the records. This review provides a comprehensive overview and framework of smartphone-based dietary assessment tools to help researchers identify suitable assessment tools for their studies. Future research needs to address the potential impact of specific dietary assessment methods on data quality and participants' willingness to record their behaviour to ultimately improve the quality of smartphone-based dietary assessment for health research.

Relative Validation of an Artificial Intelligence–Enhanced, Image-Assisted Mobile App for Dietary Assessment in Adults: Randomized Crossover Study

Background

Thorough dietary assessment is essential to obtain accurate food and nutrient intake data yet challenging because of the limitations of current methods. Image-based methods may decrease energy underreporting and increase the validity of self-reported dietary intake. Keenoa is an image-assisted food diary that integrates artificial intelligence food recognition. We hypothesized that Keenoa is as valid for dietary assessment as the automated self-administered 24-hour recall (ASA24)–Canada and better appreciated by users.

Objective

We aimed to evaluate the relative validity of Keenoa against a 24-hour validated web-based food recall platform (ASA24) in both healthy individuals and those living with diabetes. Secondary objectives were to compare the proportion of under- and overreporters between tools and to assess the user’s appreciation of the tools.

Methods

We used a randomized crossover design, and participants completed 4 days of Keenoa food tracking and 4 days of ASA24 food recalls. The System Usability Scale was used to assess perceived ease of use. Differences in reported intakes were analyzed using 2-tailed paired t tests or Wilcoxon signed-rank test and deattenuated correlations by Spearman coefficient. Agreement and bias were determined using the Bland-Altman test. Weighted Cohen κ was used for cross-classification analysis. Energy underreporting was defined as a ratio of reported energy intake to estimated resting energy expenditure <0.9.

Results

A total of 136 participants were included (mean 46.1, SD 14.6 years; 49/136, 36% men; 31/136, 22.8% with diabetes). The average reported energy intakes (kcal/d) were 2171 (SD 553) in men with Keenoa and 2118 (SD 566) in men with ASA24 (P=.38) and, in women, 1804 (SD 404) with Keenoa and 1784 (SD 389) with ASA24 (P=.61). The overall mean difference (kcal/d) was −32 (95% CI −97 to 33), with limits of agreement of −789 to 725, indicating acceptable agreement between tools without bias. Mean reported macronutrient, calcium, potassium, and folate intakes did not significantly differ between tools. Reported fiber and iron intakes were higher, and sodium intake lower, with Keenoa than ASA24. Intakes in all macronutrients (r=0.48-0.73) and micronutrients analyzed (r=0.40-0.74) were correlated (all P<.05) between tools. Weighted Cohen κ scores ranged from 0.30 to 0.52 (all P<.001). The underreporting rate was 8.8% (12/136) with both tools. Mean System Usability Scale scores were higher for Keenoa than ASA24 (77/100, 77% vs 53/100, 53%; P<.001); 74.8% (101/135) of participants preferred Keenoa.

Conclusions

The Keenoa app showed moderate to strong relative validity against ASA24 for energy, macronutrient, and most micronutrient intakes analyzed in healthy adults and those with diabetes. Keenoa is a new, alternative tool that may facilitate the work of dietitians and nutrition researchers. The perceived ease of use may improve food-tracking adherence over longer periods.

Using Popular Foods Consumed to Inform Development of Digital Tools for Dietary Assessment and Monitoring

Knowing the type and quality of the most popular foods consumed by a population can be useful in the design of technologies for monitoring food intake and interventions. The aim of this research was to determine the most frequently consumed foods and beverages among the Australian population and provide recommendations for progressing the design of dietary assessment technologies. Analysis of the first 24 h recall of the most recent Australian National Nutrition and Physical Activity Survey was conducted. The most popular foods and beverages consumed by energy (kJ) and by frequency were calculated. There were 4515 separate foods and beverages reported by 12,153 people. Overall, the top 10 foods that contributed most energy included full fat milk, beer, white rice, white bread, red wine, cola soft drinks, bananas, red apples, wholewheat breakfast cereal and white sugar. The five most frequently reported foods and beverages were tap water, black tea, full fat milk, instant coffee, and sugar. Understanding the most popular foods and beverages consumed can support innovations in the design of digital tools for dietary surveillance and to reduce under-reporting and food omissions. These findings could also guide the development of more tailored and relevant food databases that underpin these technologies.

A Narrative Review of Dietary Assessment Tools for Preschool-Aged Children in the Home Environment

Preschool-aged children in the U.S. have suboptimal diets. Interventions to improve child nutrition focus on parents and their role in shaping social and physical home environments, which influence children's eating behaviors. Dietary assessment tools selected to measure intervention objectives, and how results are interpreted in key findings, are essential when examining children's diets. The objectives of this review were to (1) describe dietary assessment tools used in intervention studies in young children focused within the home environment; and (2) examine how the application of these dietary assessment tools addressed intervention objectives. PubMed and Web of Science were searched for English-language nutrition intervention studies that included children aged 2-5 years, had a home environment component, used a dietary assessment tool, and reported on diet-related outcomes. Seventeen studies were included. Intervention objectives focused on overall diet, specific food groups, eating occasions, and obesity prevention/treatment. Concordance of key findings with intervention objectives, type of tool used, and multiple tools within the same study varied with 8 studies aligning in objective and tool, 1 discordant in both, and 8 partially concordant or too broad to determine. This review highlights current challenges in measuring dietary intake in preschoolers and provides recommendations for alternative applications and strategies.

Evaluation of paper-based and web-based food frequency questionnaires for 7-year-old children in Singapore

Advances in technology enabled the development of a web-based, pictorial FFQ to collect parent-report dietary intakes of 7-year-old children in the Growing Up in Singapore Towards healthy Outcomes study. This study aimed to compare intakes estimated from a paper-FFQ and a web-FFQ and examine the relative validity of both FFQ against 3-d diet records (3DDR). Ninety-two mothers reported food intakes of their 7-year-old child on a paper-FFQ, a web-FFQ and a 3DDR. A usability questionnaire collected participants' feedback on the web-FFQ. Correlations and agreement in energy, nutrients and food groups intakes between the dietary assessments were evaluated using Pearson's correlation, Lin's concordance, Bland-Altman plots, Cohen's κ and tertile classification. The paper- and web-FFQ had good correlations (≥ 0·50) and acceptable-good agreement (Lin's concordance ≥ 0·30; Cohen's κ ≥ 0·41; ≥ 50 % correct and ≤ 10 % misclassification into same or extreme tertiles). Compared with 3DDR, both FFQ showed poor agreement (< 0·30) in assessing absolute intakes except micronutrients (web-FFQ had acceptable-good agreement), but showed acceptable-good ability to classify children into tertiles (κ ≥ 0·21; ≥ 40 % and ≤ 15 % correct or misclassification). Bland-Altman plots suggest good agreement between web-FFQ and 3DDR in assessing micronutrients and several food groups. The web-FFQ was well-received, and majority (81 %) preferred the web-FFQ over the paper-FFQ. The newly developed web-FFQ produced intake estimates comparable to the paper-FFQ, has acceptable-good agreement with 3DDR in assessing absolute micronutrients intakes and has acceptable-good ability to classify children according to categories of intakes. The positive acceptance of the web-FFQ makes it a feasible tool for future dietary data collection.

Improved Wearable Devices for Dietary Assessment Using a New Camera System

An unhealthy diet is strongly linked to obesity and numerous chronic diseases. Currently, over two-thirds of American adults are overweight or obese. Although dietary assessment helps people improve nutrition and lifestyle, traditional methods for dietary assessment depend on self-report, which is inaccurate and often biased. In recent years, as electronics, information, and artificial intelligence (AI) technologies advanced rapidly, image-based objective dietary assessment using wearable electronic devices has become a powerful approach. However, research in this field has been focused on the developments of advanced algorithms to process image data. Few reports exist on the study of device hardware for the particular purpose of dietary assessment. In this work, we demonstrate that, with the current hardware design, there is a considerable risk of missing important dietary data owing to the common use of rectangular image screen and fixed camera orientation. We then present two designs of a new camera system to reduce data loss by generating circular images using rectangular image sensor chips. We also present a mechanical design that allows the camera orientation to be adjusted, adapting to differences among device wearers, such as gender, body height, and so on. Finally, we discuss the pros and cons of rectangular versus circular images with respect to information preservation and data processing using AI algorithms.

Clinical assessment for diet prescription

Accurate nutritional assessment based on dietary intake, physical activity, genetic makeup, and metabolites is required to prevent from developing and/or to treat people suffering from malnutrition as well as other nutrition related health issues. Nutritional screening ought to be considered as an essential part of clinical assessment for every patient on admission to healthcare setups, as well as on change in clinical conditions. Therefore, a detailed nutritional assessment must be performed every time nutritional imbalances are observed or suspected. In this review we have explored different techniques used for nutritional and physical activity assessment. Dietary Intake (DI) assessment is a multidimensional and complex process. Traditionally, dietary intake is assessed through self-report techniques, but due to limitations like biases, random errors, misestimations, and nutrient databases-linked errors, questions arise about the adequacy of self-reporting dietary intake procedures. Despite the limitations in assessing dietary intake (DI) and physical activity (PA), new methods and improved technologies such as biomarkers analysis, blood tests, genetic assessments, metabolomic analysis, DEXA (Dual-energy X-ray absorptiometry), MRI (Magnetic resonance imaging), and CT (computed tomography) scanning procedures have made much progress in the improvement of these measures. Genes also plays a crucial role in dietary intake and physical activity. Similarly, metabolites are also involved in different nutritional pathways. This is why integrating knowledge about the genetic and metabolic markers along with the latest technologies for dietary intake (DI) and physical activity (PA) assessment holds the key for accurately assessing one's nutritional status and prevent malnutrition and its related complications.

Feasibility of wearable camera use to improve the accuracy of dietary assessment among adults

Traditional methods of dietary assessment are prone to measurement error, with energy intake often under-reported. The 24-h recall is widely used in dietary assessment, however, its reliance on self-report without verification of consumption can result in inaccuracies in true nutrient intake. Wearable cameras may provide a complementary approach to improve self-report accuracy by providing an objective and passive measure of food consumption. The purpose of the present study was to determine whether a wearable camera improves the accuracy of a 24-h recall compared with a 24-h recall alone in twenty adults aged 18-65 years. The study also explored limitations associated with wearable cameras. Participants wore the camera for 1 d and a 24-h recall was then conducted the following day, before and after viewing the camera images. Dietary data were analysed using Nutritics dietary analysis software, while eating habits were assessed by a self-report questionnaire. Energy and nutrient intakes were compared between the recall alone and the camera-assisted recall. Results showed a significant increase in mean energy intake with the camera-assisted recall compared with the recall alone (9677⋅8 ± 2708⋅0 kJ/d v. 9304⋅6 ± 2588⋅5 kJ/d, respectively, P = 0⋅003). Intakes of carbohydrates, total sugars and saturated fats were also significantly higher with the camera-assisted recall. In terms of challenges, there were occasionally technological issues such as proper positioning of the camera by the participants. In conclusion, reporting of energy and nutrient intake may be enhanced when a traditional method of dietary assessment, the 24-h recall, is assisted by a wearable camera.

Association between early childhood caries and diet quality among Chinese children aged 2-5 years

Background

Early childhood caries (ECC) is a major oral problem affecting the health and wellbeing of children worldwide. Diet quality is a better predictor of ECC risk than single foods or specific nutrients. The purposes of this study were to assess the associations between ECC and diet quality among 2- to 5-year-old Chinese children.

Methods

A total of 150 eligible children were included in this study. The decayed, missing, or filled surface (dmfs) score was recorded for each child by dental examination. All participants were divided into three groups based on their age and dmfs score [the caries-free group, the ECC group, and the severe early childhood caries (S-ECC) group]. Parents were invited to complete a questionnaire on the general characteristics and oral health behaviors of the participants. The information of 24-h dietary intake from each child was captured via a mobile APP. The Chinese diet balance index for preschool children (DBI_C) indicators score, high bound score (HBS), low bound score (LBS), and diet quality distance (DQD) score were calculated to assess the diet quality of study subjects. The associations of ECC with HBS, LBS, DQD score, and DBI_C indicators score were analyzed by Mann-Whitney U test and multivariable logistic regression analysis.

Results

In this study, 21, 31, and 98 children were diagnosed with caries-free, ECC, and S-ECC, respectively. Statistical analysis revealed that the risk of ECC and S-ECC were significantly increased with the DQD score (OR = 1.283 and 1.287, respectively), but both were not associated with HBS and LBS (P > 0.05). In the meantime, the risk of ECC and S-ECC were significantly increased with the Grains score (OR = 1.623 and 1.777, respectively), and significantly decreased with the Food diversity score (OR = 0.271 and 0.315, respectively). Moreover, the risk of S-ECC also significantly decreased with the Vegetables score (OR = 0.137).

Conclusion

Both ECC and S-ECC were associated with a high degree of dietary imbalance and grains intake as well as a low degree of food diversity among Chinese children aged 2-5 years. In addition, S-ECC was also associated with a low degree of vegetable intake.