•Artificial intelligence (AI) models were generally effective for physical activity (PA) promotion (16 studies), outcome prediction (7 studies), and pattern recognition (1 study).•Twelve studies ...found AI-driven interventions, such as mobile apps, recommendation systems, and chatbots improved PA outcomes compared to traditional approaches.•An increasing trend was observed of adopting state-of-the-art deep learning and reinforcement learning models over standard machine learning.•Six key areas were identified for future AI adoption: personalized interventions, real-time monitoring and adaptation, multimodal data integration, evaluating effectiveness, expanding access, and preventing injuries.•Exploring emerging AI-driven strategies is essential for optimizing PA interventions and promoting public health.
This scoping review aimed to offer researchers and practitioners an understanding of artificial intelligence (AI) applications in physical activity (PA) interventions; introduce them to prevalent machine learning (ML), deep learning (DL), and reinforcement learning (RL) algorithms; and encourage the adoption of AI methodologies.
A scoping review was performed in PubMed, Web of Science, Cochrane Library, and EBSCO focusing on AI applications for promoting PA or predicting related behavioral or health outcomes. AI methodologies were summarized and categorized to identify synergies, patterns, and trends informing future research. Additionally, a concise primer on predominant AI methodologies within the realm of PA was provided to bolster understanding and broader application.
The review included 24 studies that met the predetermined eligibility criteria. AI models were found effective in detecting significant patterns of PA behavior and associations between specific factors and intervention outcomes. Most studies comparing AI models to traditional statistical approaches reported higher prediction accuracy for AI models on test data. Comparisons of different AI models yielded mixed results, likely due to model performance being highly dependent on the dataset and task. An increasing trend of adopting state-of-the-art DL and RL models over standard ML was observed, addressing complex human–machine communication, behavior modification, and decision-making tasks. Six key areas for future AI adoption in PA interventions emerged: personalized PA interventions, real-time monitoring and adaptation, integration of multimodal data sources, evaluation of intervention effectiveness, expanding access to PA interventions, and predicting and preventing injuries.
The scoping review highlights the potential of AI methodologies for advancing PA interventions. As the field progresses, staying informed and exploring emerging AI-driven strategies is essential for achieving significant improvements in PA interventions and fostering overall well-being.
Display omitted
Although COVID-19 has disproportionately affected socio-economically vulnerable populations, research on its impact on socio-economic disparities in unhealthy food reliance remains scarce.
This study ...uses mobile phone data to evaluate the impact of COVID-19 on socio-economic disparities in reliance on convenience stores and fast food. Reliance is defined in terms of the proportion of visits to convenience stores out of the total visits to both convenience and grocery stores, and the proportion of visits to fast food restaurants out of the total visits to both fast food and full-service restaurants. Visits to each type of food outlet at the county level were traced and aggregated using mobile phone data before being analyzed with socio-economic demographics and COVID-19 incidence data.
Our findings suggest that a new COVID-19 case per 1,000 population decreased a county's odds of relying on convenience stores by 3.41% and increased its odds of fast food reliance by 0.72%. As a county's COVID-19 incidence rate rises by an additional case per 1,000 population, the odds of relying on convenience stores increased by 0.01%, 0.02%, and 0.06% for each additional percentage of Hispanics, college-educated residents, and every additional year in median age, respectively. For fast food reliance, as a county's COVID-19 incidence rate increases by one case per 1,000 population, the odds decreased by 0.003% for every additional percentage of Hispanics but increased by 0.02% for every additional year in the county's median age.
These results complement existing literature to promote equitable food environments.
Due to the role that sugar-sweetened beverages (SSBs) play in the obesity epidemic, SSB taxes have been enacted in the United States in the California cities of Albany, Berkeley, Oakland, and San ...Francisco, as well as in Boulder, Philadelphia, and Seattle. We pooled five years of Nielsen Consumer Panel and Retail Scanner Data (2014-18) to examine purchasing behaviors in and around these cities that have instituted SSB taxes. We included households that were either subject to the tax during the study period or were in surrounding areas within the same state. The goal was to test for the differential impact of SSB taxes by income level and type of tax. Multivariate analyses of beverage purchases found that (1) there is a dose-response relationship with the size of the SSB tax; (2) the Philadelphia tax, which is the only one that includes low-calorie beverages, is associated with greater reductions in SSB purchases and an increase in bottled water purchase; and (3) approximately 72% of the tax is passed through to consumers, but this does not vary by income level of the household. Few income-related effects were detected. Overall, our findings suggest that the Philadelphia model may be the most effective at encouraging healthy habits in beverage choice.
Study objective Older adults are frequently hospitalized from the emergency department (ED) after an episode of unexplained syncope. Current admission patterns are costly, with little evidence of ...benefit. We hypothesize that an ED observation syncope protocol will reduce resource use without adversely affecting patient-oriented outcomes. Methods This randomized trial at 5 EDs compared an ED observation syncope protocol to inpatient admission for intermediate-risk adults (≥50 years) presenting with syncope or near syncope. Primary outcomes included inpatient admission rate and length of stay. Secondary outcomes included 30-day and 6-month serious outcomes after hospital discharge, index and 30-day hospital costs, 30-day quality-of-life scores, and 30-day patient satisfaction. Results Study staff randomized 124 patients. Observation resulted in a lower inpatient admission rate (15% versus 92%; 95% confidence interval CI difference −88% to −66%) and shorter hospital length of stay (29 versus 47 hours; 95% CI difference −28 to −8). Serious outcome rates after hospital discharge were similar for observation versus admission at 30 days (3% versus 0%; 95% CI difference −1% to 8%) and 6 months (8% versus 10%; 95% CI difference −13% to 9%). Index hospital costs in the observation group were $629 (95% CI difference −$1,376 to −$56) lower than in the admission group. There were no differences in 30-day quality-of-life scores or in patient satisfaction. Conclusion An ED observation syncope protocol reduced the primary outcomes of admission rate and hospital length of stay. Analyses of secondary outcomes suggest reduction in index hospital costs, with no difference in safety events, quality of life, or patient satisfaction. Our findings suggest that an ED observation syncope protocol can be replicated and safely reduce resource use.
Abstract Background The sweeping obesity epidemic could further increase the incidence of functional limitations in the U.S. rapidly aging population. Objective To examine the relationship between ...body weight status and onset of functional limitations in U.S. middle-aged and older adults. Methods Study sample came from 1992 to 2010 waves of the Health and Retirement Study, a nationally representative longitudinal survey of community-dwelling middle-aged and older adults. Body mass index (BMI) was calculated from self-reported height/weight. Functional limitations were classified into physical mobility limitation (PM), large muscle function limitation (LMF), activities of daily living limitation (ADL), gross motor function limitation (GMF), and fine motor function limitation (FMF). Mixed-effect logistic regressions were performed to estimate the relationship between prior-wave body weight status and current-wave onset of functional limitations, adjusted for individual characteristics and survey design. Results Prior-wave body weight status prospectively predicted onset of functional limitation, and the relationship showed a U-shaped pattern. Compared with their normal weight counterparts, the odds ratios (ORs) in underweight (BMI < 18.5) and obese (BMI ≥ 30) adults were 1.30 (95% confidence interval, 1.05–1.62) and 2.31 (2.11–2.52) for PM, 1.20 (0.96–1.50) and 1.63 (1.49–1.79) for LMF, 2.02 (1.66–2.46) and 1.40 (1.28–1.54) for ADL, 1.96 (1.60–2.39) and 1.77 (1.62–1.93) for GMF, and 1.66 (1.37–2.02) and 1.34 (1.22–1.46) for FMF, respectively. For PM, LMF and GMF, the impact of obesity appeared more pronounced in women, whereas that of underweight more pronounced in men. Conclusions Proper weight management during aging is crucial in preventing functional limitations in middle-aged and older adults.
Nuts are nutrient-dense foods and can be incorporated into a healthy diet. Artificial intelligence-powered diet-tracking apps may promote nut consumption by providing real-time, accurate nutrition ...information but depend on data and model availability. Our team developed a dataset comprising 1380 photographs, each in RGB color format and with a resolution of 4032 × 3024 pixels. These images feature 11 types of nuts that are commonly consumed. Each photo includes three nut types; each type consists of 2-4 nuts, so 6-9 nuts are in each image. Rectangular bounding boxes were drawn using a visual geometry group (VGG) image annotator to facilitate the identification of each nut, delineating their locations within the images. This approach renders the dataset an excellent resource for training models capable of multi-label classification and object detection, as it was meticulously divided into training, validation, and test subsets. Utilizing transfer learning in Python with the IceVision framework, deep neural network models were adeptly trained to recognize and pinpoint the nuts depicted in the photographs. The ultimate model exhibited a mean average precision of 0.7596 in identifying various nut types within the validation subset and demonstrated a 97.9% accuracy rate in determining the number and kinds of nuts present in the test subset. By integrating specific nutritional data for each type of nut, the model can precisely (with error margins ranging from 0.8 to 2.6%) calculate the combined nutritional content-encompassing total energy, proteins, carbohydrates, fats (total and saturated), fiber, vitamin E, and essential minerals like magnesium, phosphorus, copper, manganese, and selenium-of the nuts shown in a photograph. Both the dataset and the model have been made publicly available to foster data exchange and the spread of knowledge. Our research underscores the potential of leveraging photographs for automated nut calorie and nutritional content estimation, paving the way for the creation of dietary tracking applications that offer real-time, precise nutritional insights to encourage nut consumption.
Menu labeling regulations in the United States mandate chain restaurants to display calorie information for standard menu items, intending to facilitate healthy dietary choices and address obesity ...concerns. For this study, we utilized machine learning techniques to conduct a novel sentiment analysis of public opinions regarding menu labeling regulations, drawing on Twitter data from 2008 to 2022. Tweets were collected through a systematic search strategy and annotated as positive, negative, neutral, or news. Our temporal analysis revealed that tweeting peaked around major policy announcements, with a majority categorized as neutral or news-related. The prevalence of news tweets declined after 2017, as neutral views became more common over time. Deep neural network models like RoBERTa achieved strong performance (92% accuracy) in classifying sentiments. Key predictors of tweet sentiments identified by the random forest model included the author's followers and tweeting activity. Despite limitations such as Twitter's demographic biases, our analysis provides unique insights into the evolution of perceptions on the regulations since their inception, including the recent rise in negative sentiment. It underscores social media's utility for continuously monitoring public attitudes to inform health policy development, execution, and refinement.
As the Chat Generative Pre-trained Transformer (ChatGPT) achieves increased proficiency in diverse language tasks, its potential implications for academic integrity and plagiarism risks have become ...concerning. Traditional plagiarism detection tools primarily analyze text passages, which may fall short when identifying machine-generated text. This study aims to introduce a method that uses both prompts and essays to differentiate between machine-generated and human-written text, with the goal of enhancing classification accuracy and addressing concerns of academic integrity. Leveraging a dataset of student-written essays responding to eight distinct prompts, we generated comparable essays with ChatGPT. Similarity scores within machine-generated essays (“within” scores) and between human-written and machine-generated essays (“between” scores) were computed. Subsequently, we used the percentile scores of the “between” scores within the “within” scores distribution to gauge the probability of an essay being machine-generated. Our proposed method achieved high classification accuracy, with an AUC score of 0.991, a false positive rate of 0.01, and a false negative rate of 0.037 in the test set. This validates its effectiveness in distinguishing between machine-generated and human-written essays and shows that it outperforms existing approaches based solely on text passages. This research presents a straightforward and effective method to detect machine-generated essays using prompts, providing a reliable solution to maintain academic integrity in the era of advanced language models like ChatGPT. Nevertheless, the method is not without its limitations, warranting further research to investigate its performance across diverse educational contexts, various prompts, and different model hyperparameters.
: MicroRNAs have altered expression levels in various diseases and may play an important role in the diagnosis and prognosis of colorectal cancer (CRC).
: We systemically reviewed and quantitatively ...synthesized the scientific evidence pertaining to microRNA-20a (miR-20a) as a CRC biomarker. A keyword and reference search in PubMed yielded 32 studies, in which miR-20a was measured in feces, serum, or tumor tissue. Data were extracted from a total of 5014 cancer cases and 2863 controls.
: Twenty out of 21 relevant studies found that miR-20a was upregulated in CRC patients compared to controls. Meta-analysis revealed a pooled miR-20a fold change of 2.45 (95% CI: 2.24-2.66) in CRC patients versus controls. To estimate sensitivity and specificity of miR-20a as a diagnostic biomarker of CRC, a pooled area under the receiver operating characteristic curve (AUROC) was calculated (0.70, 95% CI: 0.63-0.78). The prognostic capacity of miR-20a was assessed using hazard ratios (HRs) for the overall survival (OS). The meta-analysis estimated the pooled HR for OS to be 2.02 (95% CI: 0.90-3.14) in CRC patients with high miR-20a expression.
: miR-20a may be a valid biomarker for CRC detection but may not be a strong predictor of poor prognosis in CRC.