We present a low-cost, smartwatch-based system for passive detection of cigarette smoking. It uses data from the motion sensors in the watch to identify the signature hand movements of cigarette smoking. The system will provide the detailed measures of individual smoking behaviour needed for context-triggered just-in-time smoking cessation support systems, and to enable just-in-time adaptive interventions. More broadly, the system will enable researchers to obtain detailed measures of individual smoking behaviour in free-living conditions that are free from the recall errors and reporting biases associated with self-report of smoking.

Established methods for nutritional assessment suffer from a number of important limitations. Diaries are burdensome to complete, food frequency questionnaires only capture average food intake, and both suffer from difficulties in self estimation of portion size and biases resulting from misreporting. Online and app versions of these methods have been developed, but issues with misreporting and portion size estimation remain. New methods utilizing passive data capture are required that address reporting bias, extend timescales for data collection, and transform what is possible for measuring habitual intakes. Digital and sensing technologies are enabling the development of innovative and transformative new methods in this area that will provide a better understanding of eating behavior and associations with health. In this article we describe how wrist-worn wearables, on-body cameras, and body-mounted biosensors can be used to capture data about when, what, and how much people eat and drink. We illustrate how these new techniques can be integrated to provide complete solutions for the passive, objective assessment of a wide range of traditional dietary factors, as well as novel measures of eating architecture, within person variation in intakes, and food/nutrient combinations within meals. We also discuss some of the challenges these new approaches will bring.

Smoking is associated with negative health of skin and increased signs of facial ageing. We aimed to address two questions about smoking and appearance: (1) does facial appearance alone provide an indication of smoking status, and (2) how does smoking affect the attractiveness of faces? We used faces of identical twins discordant for smoking, and prototypes made by averaging the faces of the twins. In Task 1, we presented exemplar twin sets and same sex prototypes side-by-side and participants (n = 590) indicated which face was the smoker. Participants were blind to smoking status. In Task 2 a separate sample (n = 580) indicated which face was more attractive. For the exemplar twin sets, there was inconclusive evidence participants selected the smoking twin as the smoker more often, or selected the non-smoking twin as the more attractive more often. For the prototypes, however, participants clearly selected the smoking prototypes as the smoker more often, and the non-smoking prototypes as the more attractive. Prototypical faces of smokers are judged more attractive and correctly identified as smokers more often than prototypical faces of matched non-smokers. We discuss the possible use of these findings in smoking behaviour change interventions.

Introduction: Passive detection of cigarette smoking offers potential for considerable benefits to researchers exploring smoking behaviour and designing precision behaviour change interventions. A number of systems have been developed that either use bespoke sensing technology, or rely on connected smartphones to run analytical software. Here we present StopWatch, a system for passive detection of cigarette smoking that runs on a smartwatch and does not require additional sensing or a connected smartphone.Methods: Our system uses motion data from the accelerometer and gyroscope in an Android smartwatch to detect the signature hand movements of cigarette smoking. It uses a three-stage analytical pipeline to transform raw motion data into motion features, and in turn into individual drags and instances of smoking. This pipeline runs on the smartwatch, and does not require a smartphone.Results: We validated the system in daily smokers (n=13) in laboratory and free-living conditions running on an Android LG G-Watch. In free-living conditions, over a 24 hour period, the system achieved precision of 86% and recall of 71%.Conclusions: StopWatch is a system for passive measurement of cigarette smoking that runs entirely on a commercially available smartwatch. It runs on an Android smartwatch and requires no smartphone so the cost is low. No bespoke sensing equipment is needed, and it uses a mass-market smartwatch, so participant burden is low. Performance is currently lower than other more expensive and complex systems, though adequate for some applications. Future developments will focus on enhancing performance, and validation on a range of smartwatches.

This project takes the stopWatch system for passive detection of cigarette smoking which was originally validated on an early type of smartwatch and re-validates it on a device more typical of the current generation of smartwatches.

Background: A number of different systems are available for passive detection of cigarette smoking, but few studies have reported the feasibility of using these in free-living conditions, and none so far have reported specifically on the feasibility of using these in workplace settings. Methods: We conducted a feasibility study of using stopWatch, a smartwatch-based system for passive detection of cigarette smoking, in workers in the construction industry. Participants wore stopWatch for three days midweek at work. Some also wore for three days over a weekend at home. They also kept paper diaries of cigarettes smoked. Results: Six cigarette smokers and two vapers were recruited. Mean number of cigarettes smoked per day was 6.1 and stopWatch detected on average 31% of these. Insufficient data were available for meaningful comparison of performance at work and home. No occurrences of vaping were detected as cigarette smoking by stopWatch. Discussion: The percentage of cigarettes smoked detected by stopWatch was lower than previously reported in free-living conditions (71%). We identified a number of practical reasons for this, including not keeping the smartwatch battery properly charged, the stopWatch application not being restarted correctly after the battery ran flat, and participants not wearing the smartwatch correctly. We make recommendations for addressing these issues.Conclusion: This is the first study of the feasibility of using a system for passive detection of cigarette smoking in a workplace setting. Several practical issues have been identified and recommendations made for improving the use of systems of this kind in future studies.

Established methods for nutritional assessment suffer from a number of important limitations. Diaries are burdensome to complete, food frequency questionnaires only capture average food intake, and both suffer from difficulties in self estimation of portion size and biases resulting from misreporting. Online and app versions of these methods have been developed, but issues with misreporting and portion size estimation remain. New methods utilising passive data capture are required that address reporting bias, extend timescales for data collection, and transform what is possible for measuring habitual intakes. Digital and sensing technologies are enabling the development of innovative and transformative new methods in this area that will provide a better understanding of eating behaviour and associations with health. In this article we describe how wrist-worn wearables, on-body cameras, and body-mounted biosensors can be used to capture data about when, what and how much people eat and drink. We illustrate how these new techniques can be integrated to provide complete solutions for the passive, objective assessment of a wide range of traditional dietary factors, as well as novel measures of eating architecture, within person variation in intakes, and food/nutrient combinations within meals. We also discuss some of the challenges these new approaches will bring.

The penultimate sentence in the abstract is incorrect. It currently reads as follows:'Prototypical faces of smokers are judged more attractive and correctly identified as smokers more often than prototypical faces of matched non-smokers.'The correct sentence is:'Prototypical faces of non-smokers are judged more attractive, and prototypical faces of smokers are correctly identified as smokers more often than prototypical faces of matched smokers/nonsmokers.'In Section 3.2, the last sentence should read: 'Bayesian analyses found extreme evidence to support the hypothesis that participants found the non-smoking twins more attractive (BF 10 males = 1.05e + 22, BF 10 females = 1.86e + 24).'In Section 3.4, the second paragraph should read: 'Task 2: Exact binomial tests indicated male participants judged the male non-smoking prototype (mean response = 0.28, corresponding to 72%) and the female non-smoking prototype (0.34, 66%) as more attractive, and female participants judged the male nonsmoking prototype (0.32, 68%) and the female non-smoking prototype (0.30, 70%) to be more attractive, all ps < 0.001.'