Perspective: Big Data and Machine Learning Could Help Advance Nutritional Epidemiology

Abstract

The field of nutritional epidemiology faces challenges posed by measurement error, diet as a complex exposure, and residual confounding. The objective of this perspective article is to highlight how developments in big data and machine learning can help address these challenges. New methods of collecting 24-h dietary recalls and recording diet could enable larger samples and more repeated measures to increase statistical power and measurement precision. In addition, use of machine learning to automatically classify pictures of food could become a useful complimentary method to help improve precision and validity of dietary measurements. Diet is complex due to thousands of different foods that are consumed in varying proportions, fluctuating quantities over time, and differing combinations. Current dietary pattern methods may not integrate sufficient dietary variation, and most traditional modeling approaches have limited incorporation of interactions and nonlinearity. Machine learning could help better model diet as a complex exposure with nonadditive and nonlinear associations. Last, novel big data sources could help avoid unmeasured confounding by offering more covariates, including both omics and features derived from unstructured data with machine learning methods. These opportunities notwithstanding, application of big data and machine learning must be approached cautiously to ensure quality of dietary measurements, avoid overfitting, and confirm accurate interpretations. Greater use of machine learning and big data would also require substantial investments in training, collaborations, and computing infrastructure. Overall, we propose that judicious application of big data and machine learning in nutrition science could offer new means of dietary measurement, more tools to model the complexity of diet and its relations with diseases, and additional potential ways of addressing confounding.

Exploration of Machine Learning and Statistical Techniques in Development of a Low-Cost Screening Method Featuring the Global Diet Quality Score for Detecting Prediabetes in Rural India

“…ML is beginning to find its way into nutrition studies ( 17 , 18 ), and in practice can be broadly characterized by several cornerstone concepts ( 19 ). First, a modeling approach or algorithm is selected based on the type of data under study, scientific question at hand, and relevant data and modeling assumptions of importance.…”

Prospects and Pitfalls of Machine Learning in Nutritional Epidemiology

“…Nutritional digital data can be generated through multiple means, thanks to the ubiquity of Internet-connected computers and smartphones. For example, ML models can now successfully leverage the entire contents of the electronic health records (Morgenstern et al [19]). In wearable technology (Phillips et al [20]), dietary assessment can be conducted by using wearable devices containing gyroscopes and/or accelerometers.…”

“…In (Santos et al [62]), PCA was applied on 34 variables expressing the mean food intake of 1102 individuals from a population-based study. Nonlinear dimensionality reduction techniques are deep autoencoders and manifold learning (Morgenstern et al [19]). Deep autoencoders are especially interesting since they are DNNs and combine the advantages of being both unsupervised and nonlinear approaches, allowing for the embedding of data into a low-dimensional representation while conserving its properties (Falissard et al [63], Wang et al [64]).…”

“…That is because such models have the capability of processing data through their several layers and create entirely new types of variables. This approach not only adjusts for confounding, but also for any existing nonlinearity (Morgenstern et al [19]) by taking advantage of the data richness in number of observations and variables. High-capacity models of this type however are prone to overfitting and errors in the testing phase.…”

Determining the effective factors in predicting diet adherence using an intelligent model

“…This procedure continues until the algorithm's termination condition is identified and satisfied. For example, It is possible to provide a subset of suitable factors for predicting diseases and nutrition-related problems when dealing with large data sets such as diet data 29 .…”