Leading dietary determinants identified using machine learning techniques and a healthy diet score for changes in cardiometabolic risk factors in children: a longitudinal analysis

Advances in methods for characterizing dietary patterns: A scoping review

There is a growing focus on better understanding the complexity of dietary patterns and how they relate to health and other factors. Approaches that have not traditionally been applied to characterize dietary patterns, such as machine learning algorithms and latent class analysis methods, may offer opportunities to measure and characterize dietary patterns in greater depth than previously considered. However, there has not been a formal examination of how this wide range of approaches has been applied to characterize dietary patterns. This scoping review synthesized literature from 2005-2022 applying methods not traditionally used to characterize dietary patterns, referred to as novel methods. MEDLINE, CINAHL, and Scopus were searched using keywords including machine learning, latent class analysis, and least absolute shrinkage and selection operator (LASSO). Of 5274 records identified, 24 met the inclusion criteria. Twelve of 24 articles were published since 2020. Studies were conducted across 17 countries. Nine studies used approaches that have applications in machine learning to identify dietary patterns. Fourteen studies assessed associations between dietary patterns that were characterized using novel methods and health outcomes, including cancer, cardiovascular disease, and asthma. There was wide variation in the methods applied to characterize dietary patterns and in how these methods were described. The extension of reporting guidelines and quality appraisal tools relevant to nutrition research to consider specific features of novel methods may facilitate complete and consistent reporting and enable evidence synthesis to inform policies and programs aimed at supporting healthy dietary patterns.

Role of Artificial Intelligence in Multinomial Decisions and Preventative Nutrition in Alzheimer's Disease

Alzheimer's disease (AD) affects 50 million people worldwide, an increase of 35 million since 2015, and it is known for memory loss and cognitive decline. Considering the morbidity associated with AD, it is important to explore lifestyle elements influencing the chances of developing AD, with special emphasis on nutritional aspects. This review will first discuss how dietary factors have an impact in AD development and the possible role of Artificial Intelligence (AI) and Machine Learning (ML) in preventative care of AD patients through nutrition. The Mediterranean-DASH diets provide individuals with many nutrient benefits which assists the prevention of neurodegeneration by having neuroprotective roles. Lack of micronutrients, protein-energy, and polyunsaturated fatty acids increase the chance of cognitive decline, loss of memory, and synaptic dysfunction among others. ML software has the ability to design models of algorithms from data introduced to present practical solutions that are accessible and easy to use. It can give predictions for a precise medicine approach to evaluate individuals as a whole. There is no doubt the future of nutritional science lies on customizing diets for individuals to reduce dementia risk factors, maintain overall health and brain function.

Artificial intelligence in preventive cardiology

Artificial intelligence (AI) is a field of study that strives to replicate aspects of human intelligence into machines. Preventive cardiology, a subspeciality of cardiovascular (CV) medicine, aims to target and mitigate known risk factors for CV disease (CVD). AI's integration into preventive cardiology may introduce novel treatment interventions and AI-centered clinician assistive tools to reduce the risk of CVD. AI's role in nutrition, weight loss, physical activity, sleep hygiene, blood pressure, dyslipidemia, smoking, alcohol, recreational drugs, and mental health has been investigated. AI has immense potential to be used for the screening, detection, and monitoring of the mentioned risk factors. However, the current literature must be supplemented with future clinical trials to evaluate the capabilities of AI interventions for preventive cardiology. This review discusses present examples, potentials, and limitations of AI's role for the primary and secondary prevention of CVD.

Diet Quality and Cardiometabolic Risk Factors in Adolescents with Down Syndrome

BACKGROUND

Youth with Down syndrome (DS) have a high prevalence of obesity and dyslipidemia. Diet quality may influence cardiometabolic risk (CMR) in youth.

OBJECTIVE

The aim of this secondary analysis was to investigate the relationship between diet quality (Healthy Eating Index [HEI-2015]) with CMR factors in youth with DS compared with age, sex, race, ethnicity, and body mass index percentile matched, typically developing controls.

DESIGN

Adolescents (aged 10 to 20 years) with DS and controls of comparable age, sex, race, ethnicity, and body mass index percentile were recruited from 2012 to 2017 for a cross-sectional study from two large children's hospitals (Children's Hospital of Philadelphia and the Children's National Health System in Washington, DC).

PARTICIPANTS AND SETTING

CMRs in 143 adolescents with DS were compared with 100 controls. Exclusion criteria consisted of major organ-system illnesses.

MAIN OUTCOME MEASURES

The average of three 24-hour dietary recalls was used to calculate the HEI-2015. Anthropometrics, blood pressure, and fasting labs were collected.

STATISTICAL ANALYSES PERFORMED

Group differences were tested using Wilcoxon rank-sum tests. Relationships of CMR factors with HEI-2015 score within DS and controls were tested using linear regression models adjusted for sex, age, race, and body mass index z score.

RESULTS

Compared with controls (n = 100, median age = 14.8 years [interquartile range = 12.2 to 17.3 years]; 41% male; 24% African American; 65% with body mass index ≥85th percentile), adolescents with DS (n = 143, median age = 14.7 years [interquartile range = 11.4 to 17.4 years]; 44% male; 18% African American; 62% with body mass index ≥85th percentile) had higher scores (more aligned with dietary recommendations) for total HEI-2015 (DS: 52.7 [interquartile range = 46.8 to 58.6] vs controls: 45.1 [interquartile range = 39.5 to 55.0]; P < 0.0001). Youth with DS also had higher HEI-2015 component scores for fruits, greens/beans, dairy, refined grains, and saturated fats, but lower whole grains and sodium scores. Within the group with DS, total HEI-2015 was not significantly associated with CMR measures. Whereas HEI-2015 in the DS group was negatively associated with fasting glucose levels, the difference did not meet the set level of statistical significance (-0.14, 95% CI -0.29 to 0.00; P = 0.053).

CONCLUSIONS

Adolescents in both the control and DS groups reported low-quality diets, although the DS group had HEI-2015 scores more closely aligned with recommendations. In the DS group, diet quality was not significantly associated with CMR factors. Although further research is needed, these results suggest that dyslipidemia in youth with DS may not be related to dietary intake.

Classification and Prediction on Hypertension with Blood Pressure Determinants in a Deep Learning Algorithm

Few studies classified and predicted hypertension using blood pressure (BP)-related determinants in a deep learning algorithm. The objective of this study is to develop a deep learning algorithm for the classification and prediction of hypertension with BP-related factors based on the Korean Genome and Epidemiology Study-Ansan and Ansung baseline survey. We also investigated whether energy intake adjustment is adequate for deep learning algorithms. We constructed a deep neural network (DNN) in which the number of hidden layers and the number of nodes in each hidden layer are experimentally selected, and we trained the DNN to diagnose hypertension using the dataset while varying the energy intake adjustment method in four ways. For comparison, we trained a decision tree in the same way. Experimental results showed that the DNN performs better than the decision tree in all aspects, such as having higher sensitivity, specificity, F1-score, and accuracy. In addition, we found that unlike general machine learning algorithms, including the decision tree, the DNNs perform best when energy intake is not adjusted. The result indicates that energy intake adjustment is not required when using a deep learning algorithm to classify and predict hypertension with BP-related factors.

Healthy and Sustainable Dietary Patterns in Children and Adolescents: A Systematic Review

The need for adherence to a healthy and sustainable dietary pattern in the pediatric stage is discussed worldwide, being linked to a progressive incidence of noncommunicable diseases in adulthood. The aims of this systematic review were to summarize the healthy and/or sustainable dietary patterns, defined a priori, described in the literature for use during the pediatric stage; to evaluate the adherence to these dietary patterns; and identify the health-related benefits associated with adherence to these patterns. A literature search was carried out on Medline, Scopus, and Web of Science from 2010 up to 2021, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 128 articles were included according to the following criteria: participants 2-17 y old, healthy and/or sustainable dietary patterns defined by an a priori methodology; articles written in English or Portuguese; and published since 2010. Fifty instruments with 14 adaptations that measure adherence to healthy and/or sustainable dietary patterns in children and adolescents were found. The Mediterranean Diet was the most studied dietary pattern. Adherence to healthy and/or sustainable dietary patterns has wide variations worldwide. Most of the instruments described have been little studied at pediatric ages, reducing the ability to extrapolate results. Higher adherence to these dietary patterns was associated with lower body fat, waist circumference, blood pressure, and metabolic risk. There is no consensus regarding the association with BMI. No studies have proofs of the sustainability characteristics of these instruments, it being necessary to produce a new sustainable instrument or test the association of the previous ones with, for example, the ecological footprint. Further validations of these instruments in each country and more prospective studies are needed to establish temporal relations with health-related outcomes. This systematic review was registered at www.crd.york.ac.uk/prospero/ as CRD42020221788.

Classification and Prediction on the Effects of Nutritional Intake on Overweight/Obesity, Dyslipidemia, Hypertension and Type 2 Diabetes Mellitus Using Deep Learning Model: 4-7th Korea National Health and Nutrition Examination Survey

Few studies have been conducted to classify and predict the influence of nutritional intake on overweight/obesity, dyslipidemia, hypertension and type 2 diabetes mellitus (T2DM) based on deep learning such as deep neural network (DNN). The present study aims to classify and predict associations between nutritional intake and risk of overweight/obesity, dyslipidemia, hypertension and T2DM by developing a DNN model, and to compare a DNN model with the most popular machine learning models such as logistic regression and decision tree. Subjects aged from 40 to 69 years in the 4-7th (from 2007 through 2018) Korea National Health and Nutrition Examination Survey (KNHANES) were included. Diagnostic criteria of dyslipidemia (n = 10,731), hypertension (n = 10,991), T2DM (n = 3889) and overweight/obesity (n = 10,980) were set as dependent variables. Nutritional intakes were set as independent variables. A DNN model comprising one input layer with 7 nodes, three hidden layers with 30 nodes, 12 nodes, 8 nodes in each layer and one output layer with one node were implemented in Python programming language using Keras with tensorflow backend. In DNN, binary cross-entropy loss function for binary classification was used with Adam optimizer. For avoiding overfitting, dropout was applied to each hidden layer. Structural equation modelling (SEM) was also performed to simultaneously estimate multivariate causal association between nutritional intake and overweight/obesity, dyslipidemia, hypertension and T2DM. The DNN model showed the higher prediction accuracy with 0.58654 for dyslipidemia, 0.79958 for hypertension, 0.80896 for T2DM and 0.62496 for overweight/obesity compared with two other machine leaning models with five-folds cross-validation. Prediction accuracy for dyslipidemia, hypertension, T2DM and overweight/obesity were 0.58448, 0.79929, 0.80818 and 0.62486, respectively, when analyzed by a logistic regression, also were 0.52148, 0.66773, 0.71587 and 0.54026, respectively, when analyzed by a decision tree. This study observed a DNN model with three hidden layers with 30 nodes, 12 nodes, 8 nodes in each layer had better prediction accuracy than two conventional machine learning models of a logistic regression and decision tree.

Association Between Diet Quality and Cardiometabolic Risk Factor Clustering Stratified by Socioeconomic Status Among Chinese Children

BACKGROUND

Few studies have evaluated the long-term relationship between diet quality and cardiometabolic risk factor clustering among children. The moderating effect of socio-economic status (SES) is of interest.

OBJECTIVE

To investigate the association between diet quality with cardiometabolic risk among Chinese children and to explore the moderating effect of SES.

DESIGN

In this cohort study, 5 waves (1997-2009) of the China Health and Nutrition Survey were used. Diet quality was measured by a modified version of the Chinese Children Dietary Index (mCCDI) based on Dietary Guidelines for Chinese.

PARTICIPANTS

Children between the ages of 7 and 17 (n = 2903) who completed at least 2 surveys were included. Those who missed measures or had hypertension or diabetes at baseline were excluded.

MAIN OUTCOME MEASURES

The fasting blood samples were collected in 2009. Waist circumference (WC) and blood pressure (BP) were measured in each survey.

STATISTICAL ANALYSIS PERFORMED

A continuous cardiometabolic risk score (MetScore) was derived by a confirmatory factor analysis of 5 components: WC, BP, glucose, triglycerides, and high-density lipoprotein cholesterol. Considering the latency period of the effect of behaviors, the mCCDI was lagged by the period between surveys. Linear regression was used to analyze the association of mCCDI with MetScore and its components. Mixed effect linear regression and lagged mCCDI were used for WC and BP models.

RESULTS

Higher mCCDI was independently associated with a lower MetScore at follow-up (β: -.11; 95% CI: -.18 to -.04). Higher lagged mCCDI over time was associated with a lower WC z score overall (β: -.05; 95% CI: -.08 to -.01) and among children in the low SES group (β: -.09; 95% CI: -.14 to -.04) but not those in the high SES group. When examining the 15 mCDDI components separately, scores for 5 components: more grains, vegetables, soybeans and its products; less sugar-sweetened beverages; and more diet variety were significantly associated with a lower MetScore.

CONCLUSIONS

Among Chinese children, higher diet quality measured by mCCDI was independently associated with a lower MetScore at follow-up.

Moving Health Literacy Research and Practice towards a Vision of Equity, Precision and Transparency

Over the past two decades, health literacy research has gained increasing attention in global health initiatives to reduce health disparities. While it is well-documented that health literacy is associated with health outcomes, most findings are generated from cross-sectional data. Along with the increasing importance of health literacy in policy, there is a lack of specificity and transparency about how to improve health literacy in practice. In this study, we are calling for a shift of current research paradigms from judging health literacy levels towards observing how health literacy skills are developed over the life course and practised in the real world. This includes using a life-course approach, integrating the rationale of precision public health, applying open science practice, and promoting actionable knowledge translation strategies. We show how a greater appreciation for these paradigms promises to advance health literacy research and practice towards an equitable, precise, transparent, and actionable vision.