Search

IntroductionRecent developments in Artificial Intelligence (AI) and Machine Learning (ML) technologies have opened new avenues for their applications in dietary assessments. Conventional dietary assessment methods are time-consuming, labor-driven, and have high recall bias. AI-assisted tools can be user-friendly and provide accurate dietary data. Hence, this review aimed to explore the applications of AI-assisted dietary assessment tools in real-world settings that could potentially enhance Next-Gen nutrition care delivery.Materials and methodsA total of 17,613 original, full-text articles using keywords such as “artificial intelligence OR food image analysis OR wearable devices AND dietary OR nutritional assessment,” published in English between January 2014 and September 2024 were extracted from Scopus, Web of Science, and PubMed databases. All studies exploring applications of AI-assisted dietary assessment tools with human participation were included; While methodological/developmental research and studies without human participants were excluded as this review specifically aimed to explore their applications in real-world scenarios for clinical purposes. In the final phase of screening, 66 articles were reviewed that matched our inclusion criteria and the review followed PRISMA-ScR reporting guidelines.ResultsWe observed that existing AI-assisted dietary assessment tools are integrated with mobile/web-based applications to provide a user-friendly interface. These tools can broadly be categorized as “Image-based” and “Motion sensor-based.” Image-based tools allow food recognition, classification, food volume/weight, and nutrient estimation whereas, Motion sensor-based tools help capture eating occasions through wrist movement, eating sounds, jaw motion & swallowing. These functionalities capture the dietary data regarding the type of food or beverage consumed, calorie intake, portion sizes, frequency of eating, and shared eating occasions as real-time data making it more accurate as against conventional dietary assessment methods. Dietary assessment tools integrated with AI and ML could estimate real-time energy and macronutrient intake in patients with chronic conditions such as obesity, diabetes, and dementia. Additionally, these tools are non-laborious, time-efficient, user-friendly, and provide fairly accurate data free from recall/reporting bias enabling clinicians to offer personalized nutrition.ConclusionTherefore, integrating AI-based dietary assessment tools will help improve the quality of nutrition care and navigate next-gen nutrition care practices. More studies are required further to evaluate the efficacy and accuracy of these tools.

Abstract Introduction Digital technologies have enabled new possibilities to assess dietary intake and have shown promise in terms of decreased participant burden, improved accuracy and lower costs. However, their potential and validity in pregnant populations are scarcely explored. Objectives This study aimed to (a) validate energy intakes obtained from a web-based dietary recall method developed for national surveys (RiksmatenFlex) against total energy expenditure (TEE) by means of the doubly labelled water (DLW) method, and (b) to compare intakes of macronutrients, key unhealthy and healthy foods as well as adherence to food-based dietary guidelines between RiksmatenFlex and repeated 24 h telephone dietary recalls in healthy Swedish pregnant women. Methods This study was conducted as a nested validation within the HealthyMoms trial. Intakes of foods, macronutrients and energy were assessed during three days through RiksmatenFlex and 24 h telephone dietary recalls, and Swedish Healthy Eating Index (SHEI) scores were also calculated for both methods (n = 52). For 24 women, TEE was also assessed through the DLW method. Paired Samples T-tests and Wilcoxon Signed Ranks Tests were used to identify differences between means for foods, macronutrients, energy and SHEI scores. Pearson correlation coefficient or Spearman’s rho were performed to identify relationships between variables. To compare energy intake (RiksmatenFlex) with TEE (DLW method) and 24 h telephone dietary recalls, Bland and Altman plots were constructed. Results Average energy intake from RiksmatenFlex (10,015 [SD 2004] kJ) was not statistically different from TEE (10,252 [SD 1197] kJ) (p = 0.596) (mean difference: -237 kJ/24 h). Correspondingly, there were small mean differences between average intakes of key unhealthy and healthy foods and average SHEI scores between RiksmatenFlex and 24 h telephone dietary recalls. However, the Bland and Altman plots showed wide limits of agreement for all dietary variables (e.g., for energy intake using RiksmatenFlex versus TEE: ±4239 kJ/24 h). High correlations between the investigated dietary variables for the two dietary methods were observed (r = 0.751 to 0.931; all p

Introduction: Digital technologies have enabled new possibilities to assess dietary intake and have shown promise in terms of decreased participant burden, improved accuracy and lower costs. However, their potential and validity in pregnant populations are scarcely explored. Objectives: This study aimed to (a) validate energy intakes obtained from a web-based dietary recall method developed for national surveys (RiksmatenFlex) against total energy expenditure (TEE) by means of the doubly labelled water (DLW) method, and (b) to compare intakes of macronutrients, key unhealthy and healthy foods as well as adherence to food-based dietary guidelines between RiksmatenFlex and repeated 24 h telephone dietary recalls in healthy Swedish pregnant women. Methods: This study was conducted as a nested validation within the HealthyMoms trial. Intakes of foods, macronutrients and energy were assessed during three days through RiksmatenFlex and 24 h telephone dietary recalls, and Swedish Healthy Eating Index (SHEI) scores were also calculated for both methods (n = 52). For 24 women, TEE was also assessed through the DLW method. Paired Samples T-tests and Wilcoxon Signed Ranks Tests were used to identify differences between means for foods, macronutrients, energy and SHEI scores. Pearson correlation coefficient or Spearman's rho were performed to identify relationships between variables. To compare energy intake (RiksmatenFlex) with TEE (DLW method) and 24 h telephone dietary recalls, Bland and Altman plots were constructed. Results: Average energy intake from RiksmatenFlex (10,015 [SD 2004] kJ) was not statistically different from TEE (10,252 [SD 1197] kJ) (p = 0.596) (mean difference: -237 kJ/24 h). Correspondingly, there were small mean differences between average intakes of key unhealthy and healthy foods and average SHEI scores between RiksmatenFlex and 24 h telephone dietary recalls. However, the Bland and Altman plots showed wide limits of agreement for all dietary variables (e.g., for energy intake using RiksmatenFlex versus TEE: ±4239 kJ/24 h). High correlations between the investigated dietary variables for the two dietary methods were observed (r = 0.751 to 0.931; all p < 0.001). Conclusion: RiksmatenFlex captured average intakes of energy, unhealthy and healthy food groups and adherence to food-based dietary guidelines in a comparable way to 24 h telephone dietary recalls and the DLW method. Our results support the validity of RiksmatenFlex as a web-based dietary assessment method for future use in pregnancy for intervention studies and national dietary surveys. [ABSTRACT FROM AUTHOR]

Background: Accurate measurement of food and nutrient intake is crucial for nutrition research, dietary surveillance, and disease management, but traditional methods such as 24-hour dietary recalls, food diaries, and food frequency questionnaires are often prone to recall error and social desirability bias, limiting their reliability. With the advancement of artificial intelligence (AI), there is potential to overcome these limitations through automated, objective, and scalable dietary assessment techniques. However, the effectiveness and challenges of AI applications in this domain remain inadequately explored. Objective: This study aimed to conduct a scoping review to synthesize existing literature on the efficacy, accuracy, and challenges of using AI tools in assessing food and nutrient intakes, offering insights into their current advantages and areas of improvement. Methods: This review followed the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. A comprehensive literature search was conducted in 4 databases—PubMed, Web of Science, Cochrane Library, and EBSCO—covering publications from the databases' inception to June 30, 2023. Studies were included if they used modern AI approaches to assess food and nutrient intakes in human subjects. Results: The 25 included studies, published between 2010 and 2023, involved sample sizes ranging from 10 to 38,415 participants. These studies used a variety of input data types, including food images (n=10), sound and jaw motion data from wearable devices (n=9), and text data (n=4), with 2 studies combining multiple input types. AI models applied included deep learning (eg, convolutional neural networks), machine learning (eg, support vector machines), and hybrid approaches. Applications were categorized into dietary intake assessment, food detection, nutrient estimation, and food intake prediction. Food detection accuracies ranged from 74% to 99.85%, and nutrient estimation errors varied between 10% and 15%. For instance, the RGB-D (Red, Green, Blue-Depth) fusion network achieved a mean absolute error of 15% in calorie estimation, and a sound-based classification model reached up to 94% accuracy in detecting food intake based on jaw motion and chewing patterns. In addition, AI-based systems provided real-time monitoring capabilities, improving the precision of dietary assessments and demonstrating the potential to reduce recall bias typically associated with traditional self-report methods. Conclusions: While AI demonstrated significant advantages in improving accuracy, reducing labor, and enabling real-time monitoring, challenges remain in adapting to diverse food types, ensuring algorithmic fairness, and addressing data privacy concerns. The findings suggest that AI has transformative potential for dietary assessment at both individual and population levels, supporting precision nutrition and chronic disease management. Future research should focus on enhancing the robustness of AI models across diverse dietary contexts and integrating biological sensors for a holistic dietary assessment approach. [ABSTRACT FROM AUTHOR]

For centuries, human activities have altered the population dynamics of wildlife. New anthropogenic food sources provide a predictable and abundant food supply that often induces very significant changes in the size, distribution, and behaviour of many populations, with ultimate consequences on the structure and functioning of natural ecosystems. Here, we combine historical and contemporary feather samples of a population of a superabundant, opportunistic predator, the yellow-legged gull Larus michahellis, to assess its trophic ecology and relate it to human activities in the long term. Dietary assessments were based on stable isotope analysis of carbon, nitrogen, and sulphur (δ13C, δ15N and δ34S), and were conducted through three end-point (marine prey, waste from landfills-slaughterhouses, and terrestrial invertebrates) Bayesian mixing models. Our results suggest that gulls’ diet showed a progressive decrease in the consumption of marine prey throughout the most recent period (late 20th century onwards), linked to an increase in the consumption of meat waste and small terrestrial invertebrates. Reported dietary changes over the sampling period correlated positively with the availability of marine resources around the breeding area. We provide evidence suggesting that the ability of gulls to exploit efficiently diverse anthropogenic food subsidies likely resulted in the exponential demographic increase of this population throughout the 20th century. In addition, current regulations affecting the availability of these food resources (e.g., fishing discards and landfill waste) likely reversed this trend over the last decade. Long-term evidence of population trophic plasticity, like the one we present here, is essential to implement and support management and conservation actions that limit the availability of anthropogenic resources, especially when it comes to superabundant, problematic species.

Objective: The purpose of this methodology study is to examine the relative validity of four dietary assessment methods during pregnancy and to understand the pros and cons of each method. Materials and methods: We recruited 181 healthy pregnant women with less than 20 weeks of gestation and collected information through personal and telephone interviews in Taipei, Taiwan. Dietary assessment methods including 24-hour recalls, 3-day food records, and the weekly food frequency questionnaire during the 3 trimesters and the meal-based Chinese food frequency questionnaire (CFFQ) in the 3rd trimester were used in this prospective study. Results: The percentages of energy from protein (15%), fat (31–34%), and carbohydrate (50–54%) were similar by the recall and record methods. The energy intakes from 24-hour recalls were 1924 kcal, 1980 kcal, and 2172 kcal in the 1st, 2nd, and 3rd trimesters, respectively. The weekly food frequency questionnaire resulted in significantly higher intakes of all energy nutrients, especially for protein and fat intakes (percentages and densities) for the 3 trimesters; a further adjustment for the food list is needed. The CFFQ showed comparable results with the quantitative methods in estimating dietary patterns for the entire pregnancy. The 24-hour recalls by telephone interviews provided reasonable results in the assessment of the average nutrient intakes during the trimesters. Conclusion: The combinations of 24-hour recalls for the short-term dietary changes and the CFFQ for long-term dietary patterns are suggested as appropriate dietary assessment methods during pregnancy in Taiwan.

The aim of this study was to validate the estimated habitual dietary intake from a newly developed web-based FFQ (WebFFQ), for use in an adult population in Norway. In total, ninety-two individuals were recruited. Total energy expenditure (TEE) measured by doubly labelled water was used as the reference method for energy intake (EI) in a subsample of twenty-nine women, and multiple 24-h recalls (24HR) were used as the reference method for the relative validation of macronutrients and food groups in the entire sample. Absolute differences, ratios, crude and deattenuated correlations, cross-classifications, Bland–Altman plot and plots between misreporting of EI (EI–TEE) and the relative misreporting of food groups (WebFFQ–24HR) were used to assess the validity. Results showed that EI on group level was not significantly different from TEE measured by doubly labelled water (0·7 MJ/d), but ranking abilities were poor (r−0·18). The relative validation showed an overestimation for the majority of the variables using absolute intakes, especially for the food groups ‘vegetables’ and ‘fish and shellfish’, but an improved agreement between the test and reference tool was observed for energy adjusted intakes. Deattenuated correlation coefficients were between 0·22 and 0·89, and low levels of grossly misclassified individuals (0–3 %) were observed for the majority of the energy adjusted variables for macronutrients and food groups. In conclusion, energy estimates from the WebFFQ should be used with caution, but the estimated absolute intakes on group level and ranking abilities seem acceptable for macronutrients and most food groups. [ABSTRACT FROM PUBLISHER]

Vitamin D and calcium are essential nutrients with a range of biological effects of public health relevance. This study aimed to validate a short food frequency questionnaire (SFFQ) against a three-day food record (3D record), assessing the intake of vitamin D and calcium in Swedish children during wintertime. In a double-blinded, randomized food-based intervention study on the effect of feeding different daily doses of vitamin D supplement to 5-7-year-old children (n = 85), 79 (93%) participants completed SFFQ1 at baseline and SFFQ2 after the intervention, and 72 were informed to fill in a 3D record. The 28 (39%) children who completed the 3D record were included in this validation study. The baseline level of serum-25 hydroxy vitamin D [S-25(OH)D] was used as a biomarker. The correlation between all three instruments were moderate to strong. SFFQ2 and the 3D record correlated moderately to S-25(OH)D. Bland-Altman analysis showed that SFFQ2 overestimated vitamin D intake by on average 0.6 _g/day, (limits of agreement (LOA) 5.7 and -4.6 _g/day), whereas the intake of calcium was underestimated by on average 29 mg/day, (LOA 808 and -865 mg/day). Finally, the validity coefficient calculated for vitamin D using the method of triad was high (0.75). In conclusion, this SFFQ, assessed by a dietician, is a valid tool to assess dietary vitamin D and calcium intake in groups of young children. [ABSTRACT FROM AUTHOR]

Dietary assessment in older adults can be challenging. The Novel Assessment of Nutrition and Ageing (NANA) method is a touch-screen computer-based food record that enables older adults to record their dietary intakes. The objective of the present study was to assess the relative validity of the NANA method for dietary assessment in older adults. For this purpose, three studies were conducted in which a total of ninety-four older adults (aged 65–89 years) used the NANA method of dietary assessment. On a separate occasion, participants completed a 4 d estimated food diary. Blood and 24 h urine samples were also collected from seventy-six of the volunteers for the analysis of biomarkers of nutrient intake. The results from all the three studies were combined, and nutrient intake data collected using the NANA method were compared against the 4 d estimated food diary and biomarkers of nutrient intake. Bland–Altman analysis showed a reasonable agreement between the dietary assessment methods for energy and macronutrient intake; however, there were small, but significant, differences for energy and protein intake, reflecting the tendency for the NANA method to record marginally lower energy intakes. Significant positive correlations were observed between urinary urea and dietary protein intake using both the NANA and the 4 d estimated food diary methods, and between plasma ascorbic acid and dietary vitamin C intake using the NANA method. The results demonstrate the feasibility of computer-based dietary assessment in older adults, and suggest that the NANA method is comparable to the 4 d estimated food diary, and could be used as an alternative to the food diary for the short-term assessment of an individual's dietary intake. [ABSTRACT FROM PUBLISHER]

Little is known about dietary patterns on weekdays and weekend days in children, and the aim of the present study was to investigate 4–14-year-old children's dietary patterns specifically on weekdays (Monday–Thursday) and weekend days (Saturday–Sunday). Dietary data were derived from the Danish National Survey of Dietary Habits and Physical Activity 2003–8, where a total of 784 children aged 4–14 years completed a 7 d pre-coded food record. Principal component analysis was used to identify dietary patterns in the age groups 4–6, 7–10 and 11–14 years. Consistently, two dietary patterns, labelled ‘processed’ and ‘health conscious’, emerged on both weekdays and weekend days. Factor scores from corresponding dietary patterns were significantly correlated between weekdays and weekend days with the exception of the ‘health conscious’ pattern in the 7–10-year-olds. Within each age group, children with high agreement for the ‘processed’ pattern had a significantly higher dietary energy density, which was reflected in significantly higher intakes of sugar-sweetened beverages and lower intakes of fruit and vegetables, compared with children with high agreement for the ‘health conscious’ pattern (P< 0·05). Moreover, these variables indicated less healthy dietary intakes on weekend days than on weekdays for both patterns. In conclusion, two distinct dietary patterns, labelled ‘processed’ and ‘health conscious’, were identified on both weekdays and weekend days for each age group. While overall major dietary patterns may somewhat track between weekdays and weekends, the specific foods actually eaten became less healthy during weekends. [ABSTRACT FROM PUBLISHER]

Hmong are one of the immigrant Asian subgroups with increasing rates of obesity, type 2 diabetes and CVD. Despite their population growth in the USA and declining health status, no research has investigated the appropriateness of dietary assessment measures, including FFQ and 24 h dietary recalls among Hmong. The present study compared the nutrient information obtained through a 24 h dietary recall method with that collected using the Block Kid's Food Frequency Questionnaire (Block FFQ) among Hmong-American children (n 335) of 9–18 years of age. For this purpose, two 24 h dietary recalls were collected during non-consecutive days and averaged for comparison. The Block FFQ was administered on the day of the second 24 h recall and the two methodologies were also compared using t tests. Among all children, Block FFQ nutrient estimates for vitamin A, vitamin C and food group servings for vegetables and fruits were significantly higher than those assessed through the 24 h dietary recalls (P< 0·001). Nutrient estimates for protein and food group servings for grains and meat and beans were significantly higher among all participants when assessed through the 24 h dietary recalls than through the Block FFQ (P< 0·05). The results suggest that the Block FFQ does not appear to be a good measure of protein, grain, and meat and bean intake among Hmong children of 9–18 years of age, and the 24 h dietary recall offers a better account of the Hmong diet and is inclusive of ethnic food options otherwise missed in the Block FFQ. We recommend the modification of the current Block FFQ to appropriately reflect cultural food/beverage items of the population in interest. [ABSTRACT FROM PUBLISHER]

The occurrence of 27 antibiotics (18 sulfonamides and 9 quinolones) in 92 samples of cultured Chinese mitten crab (Eriocheir sinensis) from three provinces (Anhui, Jiangsu and Liaoning) was investigated. These 21 antibiotics were detected at least once in crabs from these provinces with detection frequencies of 3.70‐90.91%. Sulfonamides were detected in 53.7% of the samples at concentrations of 0.1–10 μg/kg in Jiangsu, while quinolones were detected with 90.9% of samples containing concentrations of 1–100 μg/kg in Liaoning province. Enroflxacin, ciprofloxacin, sulfaquinoxaline, sulfameter, sulfadoxine, and sulfamethoxazole were the mainly used antibiotics and enroflxacin were present at a high concentration (>100 μg/kg). Dietary assessments showed that residual antibiotics in crabs from China were far below the maximum residue limit (MRL) of total sulfonamides and quinolones, and there was almost no risk associated with crab consumption. These results will provide meaningful indications for the safety of crab consumption. • Enroflxacin and ciprofloxacin were the main types of antibiotics used in aquaculture ponds for cultured crabs. • Dietary assessment showed that EDI of all samples accounted for less than 1.6% of the ADI. • There were almost no risk associated crab consumption in China. [ABSTRACT FROM AUTHOR]