Automated Identification of Hookahs (Waterpipes) on Instagram: An Application in Feature Extraction Using Convolutional Neural Network and Support Vector Machine Classification.

BACKGROUND: Instagram, with millions of posts per day, can be used to inform public health surveillance targets and policies. However, current research relying on image-based data often relies on hand coding of images, which is time-consuming and costly, ultimately limiting the scope of the study. Current best practices in automated image classification (eg, support vector machine (SVM), backpropagation neural network, and artificial neural network) are limited in their capacity to accurately distinguish between objects within images. OBJECTIVE: This study aimed to demonstrate how a convolutional neural network (CNN) can be used to extract unique features within an image and how SVM can then be used to classify the image. METHODS: Images of waterpipes or hookah (an emerging tobacco product possessing similar harms to that of cigarettes) were collected from Instagram and used in the analyses (N=840). A CNN was used to extract unique features from images identified to contain waterpipes. An SVM classifier was built to distinguish between images with and without waterpipes. Methods for image classification were then compared to show how a CNN+SVM classifier could improve accuracy. RESULTS: As the number of validated training images increased, the total number of extracted features increased. In addition, as the number of features learned by the SVM classifier increased, the average level of accuracy increased. Overall, 99.5% (418/420) of images classified were correctly identified as either hookah or nonhookah images. This level of accuracy was an improvement over earlier methods that used SVM, CNN, or bag-of-features alone. CONCLUSIONS: A CNN extracts more features of images, allowing an SVM classifier to be better informed, resulting in higher accuracy compared with methods that extract fewer features. Future research can use this method to grow the scope of image-based studies. The methods presented here might help detect increases in the popularity of certain tobacco products over time on social media. By taking images of waterpipes from Instagram, we place our methods in a context that can be utilized to inform health researchers analyzing social media to understand user experience with emerging tobacco products and inform public health surveillance targets and policies.

The Why We Retweet scale.

BACKGROUND: Twitter offers a platform for rapid diffusion of information and its users' attitudes and behaviors. Insights about information propagation via retweets (the message forwarding function) offer observable explanations of ways in which modern human interactions get organized in the form of online networks, and contextualized in the form of public health, policy decisions, disaster management, and civic participation. This study conceptualized and validated the Why We Retweet Scale to contextualize retweeting behavior. OBJECTIVE: Twitter users were identified using clustering algorithms that consider a users' position in their network and invited for an online survey. Participants (N = 1433) responded to 19 questions about why they retweet. Exploratory factor Analysis (EFA) was conducted on a scale development sample (70% of original sample), which informed the Confirmatory Factor Analysis (CFA) on a scale testing sample (30% of the original sample). Varimax rotation was used to obtain a rotated factor solution, which resulted in interpretable factors. Demographic differences among scale factors were analyzed using one-way ANOVA or independent samples t-tests. RESULTS: The final model (χ221 = 28, RMSEA = .03 [90% CI, 0.00-0.06], CFA = .99, TLI = 0.99) represented a parsimonious solution with 4 factors, measured by 2-3 items each, creating a final scale consisting of 9 items. Factor labels and definitions were: (1) Show approval, "Show support to the tweeter"; (2) Argue, "To argue against a tweet that I disagree with"; (3) Gain attention, "Add followers or gain attention"; and (4) Entertain, "Create humor/amusement". Demographic differences were also reported. CONCLUSIONS: The Why We Retweet Scale offers a useful conceptualization and assessment of motivations for retweeting. In the future, communication strategists might consider the factors associated with information propagation when designing campaign messages to maximize message reach and engagement on Twitter.

Employee and customer handling of nicotine-containing e-liquids in vape shops.

INTRODUCTION: Vape shops sell electronic cigarettes and related products such as e-liquids, which may contain nicotine. Direct contact with nicotine can lead to adverse health effects, and few regulations exist on how nicotine is handled in vape shops. This study examined how customers and employees come into contact with, and handle, nicotine-containing e-liquids in vape shops with the goal of informing potential future regulation of nicotine handling in vape shops. METHODS: Data were collected from 77 vape shops in the Los Angeles basin. Characteristics of the shops were documented by employee interviews and in store observations. Data collection was focused on shops located in areas with high concentrations of communities of interest; 20 shops from African-American communities, 17 from Hispanic communities, 18 from Korean communities, and 22 from non-Hispanic White communities. RESULTS: Half of the vape shops allowed customers to sample e-liquids with nicotine. Most of the shops (83%) provided self-service sampling stations for customers. A majority of shop employees (72%) reported that spills of e-liquids containing nicotine had occurred in the past. While 64% of the shops provided safety equipment, only 34% provided equipment for proper nicotine handling. Furthermore, 62% of shop employees reported handling nicotine without gloves or other safety equipment. CONCLUSIONS: Regulation on the handling of nicotine by customers and vape shop employees is important to prevent unsafe practices and subsequent injury. The frequent occurrence of spills and limited availability of safety equipment in vape shops highlights the need for the creation and enforcement of regulations to protect employees and customers. Appropriate safety training and equipment should be provided to employees to prevent accidental exposure to nicotine. Information on ways to safely handle nicotine should be communicated to vape shop employees and customers.

The protective role of school friendship ties for substance use and aggressive behaviors among middle school students.

BACKGROUND: Few studies have examined the interplay among quantifiable aspects of peer bonds, friendship types, personal characteristics, and behavioral outcomes in schools in distressed neighborhoods. The aim of this study was to identify compensatory and protective factors that can be promoted in school-based prevention programs. METHODS: The sample was comprised of students in East Los Angeles County (N=184). We investigated the association between 3 measures of social influence (friends in gangs, nominations of schoolmates as friends [out-degree], and the number of nominations received from schoolmates [in-degree]) and social self-control with lifetime alcohol, tobacco, inhalant, "other" drug use, and aggression. RESULTS: Friendships were protective for substance use and aggression and moderated the relationship between social self-control, substance use, and aggression. We found important sex differences; girls who nominated more friends were less likely to report alcohol use and aggression relative to boys but were more likely to have reported drug use as social self-control scores increased. CONCLUSIONS: Our results have important implications for school-based prevention and intervention programs. We provide preliminary evidence that school ties and perceptions of belongingness can mitigate the effects of several risk factors linked to substance use and aggression.

Problematic substance use among Hispanic adolescents and young adults: implications for prevention efforts.

Using data collected between 2005 and 2012 from a longitudinal study of acculturation patterns and substance use among Hispanic youth in Southern California (N = 2722), we fit multivariate logistic regression models to estimate the association of type and frequency of drug use, friend and parent drug use, cultural orientation (measured by the ARSMA-II), and psychological distress (CES-D score) in 10th grade with problematic substance use (measured with the RAPI) in (i) 11th grade and (ii) young adulthood. We conclude that future intervention efforts with Hispanic adolescents and young adults should target polysubstance and problem users and emphasize inter-individual, structural, and cultural processes as they relate to problematic substance use.

Substance Use Among Adolescents in California: A Latent Class Analysis.

Data from the California Healthy Kids Survey of 7th, 9th, and 11th graders were used to identify latent classes/clusters of alcohol, tobacco, and marijuana use (N = 418,702). Analyses revealed four latent classes of substance use, which included nonusers (61.1%), alcohol experimenters (some recent alcohol use; 22.8%), mild polysubstance users (lifetime use of all substances with less than 3 days of recent use; 9.2%), and frequent polysubstance users (used all substances three or more times in the past month; 6.9%). The results revealed that alcohol and marijuana use are salient to California adolescents. This information can be used to target and tailor school-based prevention efforts.