An integrative machine learning approach to discovering multi-level molecular mechanisms of obesity using data from monozygotic twin pairs

Harnessing Artificial Intelligence in Obesity Research and Management: A Comprehensive Review

Purpose: This review aims to explore the clinical and research applications of artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), in understanding, predicting, and managing obesity. It assesses the use of AI tools to identify obesity-related risk factors, predict outcomes, personalize treatments, and improve healthcare interventions for obesity. Methods: A comprehensive literature search was conducted using PubMed and Google Scholar, with keywords including "artificial intelligence", "machine learning", "deep learning", "obesity", "obesity management", and related terms. Studies focusing on AI's role in obesity research, management, and therapeutic interventions were reviewed, including observational studies, systematic reviews, and clinical applications. Results: This review identifies numerous AI-driven models, such as ML and DL, used in obesity prediction, patient stratification, and personalized management strategies. Applications of AI in obesity research include risk prediction, early detection, and individualization of treatment plans. AI has facilitated the development of predictive models utilizing various data sources, such as genetic, epigenetic, and clinical data. However, AI models vary in effectiveness, influenced by dataset type, research goals, and model interpretability. Performance metrics such as accuracy, precision, recall, and F1-score were evaluated to optimize model selection. Conclusions: AI offers promising advancements in obesity management, enabling more personalized and efficient care. While technology presents considerable potential, challenges such as data quality, ethical considerations, and technical requirements remain. Addressing these will be essential to fully harness AI's potential in obesity research and treatment, supporting a shift toward precision healthcare.

A proteogenomic analysis of the adiposity colorectal cancer relationship identifies GREM1 as a probable mediator

BACKGROUND

Adiposity is an established risk factor for colorectal cancer (CRC). The pathways underlying this relationship, and specifically the role of circulating proteins, are unclear.

METHODS

Utilizing two-sample univariable Mendelian randomization (UVMR), multivariable Mendelian randomization (MVMR), and colocalization, based on summary data from large sex-combined and sex-specific genetic studies, we estimated the univariable associations between: (i) body mass index (BMI) and waist-hip ratio (WHR) and overall and site-specific (colon, proximal colon, distal colon, and rectal) CRC risk, (ii) BMI and WHR and circulating proteins, and (iii) adiposity-associated circulating proteins and CRC risk. We used MVMR to investigate the potential mediating role of adiposity- and CRC-related circulating proteins in the adiposity-CRC association.

RESULTS

BMI and WHR were positively associated with CRC risk, with similar associations by anatomical tumor site. In total, 6591 adiposity-protein (2628 unique circulating proteins) and 33 protein-CRC (7 unique circulating proteins) associations were identified using UVMR and colocalization. One circulating protein, GREM1, was associated with BMI (only) and CRC outcomes in a manner that was consistent with a potential mediating role in sex-combined and female-specific analyses. In MVMR, adjusting the BMI-CRC association for GREM1, effect estimates were attenuated-suggestive of a potential mediating role-most strongly for the BMI-overall CRC association in women.

CONCLUSION

Results highlight the impact of adiposity on the plasma proteome and of adiposity-associated circulating proteins on the risk of CRC. Supported by evidence from UVMR and colocalization analyses using cis-single-nucleotide polymorphisms, GREM1 was identified as a potential mediator of the BMI-CRC association, particularly in women.

Obesity-related inflammatory protein signature in cardiovascular clinical outcomes: results from the Gutenberg Health Study

OBJECTIVE

The objective of this study was to investigate whether an obesity-related inflammatory protein signature (OIPS) is associated with adverse cardiovascular events.

METHODS

The Olink Target 96 Inflammation panel was performed in 6662 participants from the population-based Gutenberg Health Study (GHS). The OIPS was selected by a logistic regression model, and its association with cardiovascular outcomes was evaluated by Cox regression analysis. The GHS-derived OIPS was externally validated in the MyoVasc study.

RESULTS

The identified OIPS entailed 21 proteins involved in chemokine activity, tumor necrosis factor (TNF) receptor binding, and growth factor receptor binding. The signature revealed a novel positive association of axis inhibition protein 1 with obesity. The OIPS was associated with increased risk of all-cause and cardiac deaths, major adverse cardiovascular events, and incident coronary artery disease, independent of clinical covariates and established risk instruments. A BMI-stratified analysis confirmed the association of OIPS with increased death in those with obesity and overweight and with increased risk for coronary artery disease in those with obesity. The association of OIPS with increased risk of all-cause and cardiac deaths was validated in the MyoVasc cohort.

CONCLUSIONS

The OIPS showed a significant association with adverse clinical outcomes, particularly in those with overweight and obesity, and represents a promising tool for identifying patients at higher risk for worse cardiovascular outcomes.

Systematic Review of Machine Learning applied to the Prediction of Obesity and Overweight

Obesity and overweight has increased in the last year and has become a pandemic disease, the result of sedentary lifestyles and unhealthy diets rich in sugars, refined starches, fats and calories. Machine learning (ML) has proven to be very useful in the scientific community, especially in the health sector. With the aim of providing useful tools to help nutritionists and dieticians, research focused on the development of ML and Deep Learning (DL) algorithms and models is searched in the literature. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol has been used, a very common technique applied to carry out revisions. In our proposal, 17 articles have been filtered in which ML and DL are applied in the prediction of diseases, in the delineation of treatment strategies, in the improvement of personalized nutrition and more. Despite expecting better results with the use of DL, according to the selected investigations, the traditional methods are still the most used and the yields in both cases fluctuate around positive values, conditioned by the databases (transformed in each case) to a greater extent than by the artificial intelligence paradigm used. Conclusions: An important compilation is provided for the literature in this area. ML models are time-consuming to clean data, but (like DL) they allow automatic modeling of large volumes of data which makes them superior to traditional statistics.

Applications of Artificial Intelligence to Obesity Research: Scoping Review of Methodologies

BACKGROUND

Obesity is a leading cause of preventable death worldwide. Artificial intelligence (AI), characterized by machine learning (ML) and deep learning (DL), has become an indispensable tool in obesity research.

OBJECTIVE

This scoping review aimed to provide researchers and practitioners with an overview of the AI applications to obesity research, familiarize them with popular ML and DL models, and facilitate the adoption of AI applications.

METHODS

We conducted a scoping review in PubMed and Web of Science on the applications of AI to measure, predict, and treat obesity. We summarized and categorized the AI methodologies used in the hope of identifying synergies, patterns, and trends to inform future investigations. We also provided a high-level, beginner-friendly introduction to the core methodologies to facilitate the dissemination and adoption of various AI techniques.

RESULTS

We identified 46 studies that used diverse ML and DL models to assess obesity-related outcomes. The studies found AI models helpful in detecting clinically meaningful patterns of obesity or relationships between specific covariates and weight outcomes. The majority (18/22, 82%) of the studies comparing AI models with conventional statistical approaches found that the AI models achieved higher prediction accuracy on test data. Some (5/46, 11%) of the studies comparing the performances of different AI models revealed mixed results, indicating the high contingency of model performance on the data set and task it was applied to. An accelerating trend of adopting state-of-the-art DL models over standard ML models was observed to address challenging computer vision and natural language processing tasks. We concisely introduced the popular ML and DL models and summarized their specific applications in the studies included in the review.

CONCLUSIONS

This study reviewed AI-related methodologies adopted in the obesity literature, particularly ML and DL models applied to tabular, image, and text data. The review also discussed emerging trends such as multimodal or multitask AI models, synthetic data generation, and human-in-the-loop that may witness increasing applications in obesity research.

Novel associations between inflammation-related proteins and adiposity: A targeted proteomics approach across four population-based studies

Chronic low-grade inflammation has been proposed as a linking mechanism between obesity and the development of inflammation-related conditions such as insulin resistance and cardiovascular disease. Despite major advances in the last 2 decades, the complex relationship between inflammation and obesity remains poorly understood. Therefore, we aimed to identify novel inflammation-related proteins associated with adiposity. We investigated the association between BMI and waist circumference and 72 circulating inflammation-related proteins, measured using the Proximity Extension Assay (Olink Proteomics), in 3,308 participants of four independent European population-based studies (KORA-Fit, BVSII, ESTHER, and Bialystok PLUS). In addition, we used body fat mass measurements obtained by Dual-energy X-ray absorptiometry (DXA) in the Bialystok PLUS study to further validate our results and to explore the relationship between inflammation-related proteins and body fat distribution. We found 14 proteins associated with at least one measure of adiposity across all four studies, including four proteins for which the association is novel: DNER, SLAMF1, RANKL, and CSF-1. We confirmed previously reported associations with CCL19, CCL28, FGF-21, HGF, IL-10RB, IL-18, IL-18R1, IL-6, SCF, and VEGF-A. The majority of the identified inflammation-related proteins were associated with visceral fat as well as with the accumulation of adipose tissue in the abdomen and the trunk. In conclusion, our study provides new insights into the immune dysregulation observed in obesity that might help uncover pathophysiological mechanisms of disease development.