Are you smoking? Automatic alert system helping people keep away from cigarettes

Tobacco smoking is responsible for one out of every five deaths in the US, according to the Centers for Disease Control and Prevention (CDC). Recent advances in treatment delivery include technology-based mobile health approaches, which seek to deliver real-time feedback to smokers to aid quit attempts and mitigate lapses. With regard to the measurement of smoking, clinical trials rely on participant self-report and/or biochemical verification of smoking status to evaluate outcomes. Wearable sensors have the potential to improve current approaches by providing personalized feedback and objective verification of smoking status (Burns, 2000). In this paper, we describe the development of a novel smoking cessation system that combines motion detection and an Android software application to monitor smoking in real-time. In this system, a personalized smoking cessation plan will be created based on the goal of complete cessation or smoking reduction. Once the plan is created, the mobile system will monitor the users׳ smoking activity and provide feedback. An LSTM algorithm has been computed to train and test the motion data, which was collected from two armbands, to detect smoking and non-smoking motions. The internet message service will be used to remind users to stick to their plan when the sensor detects current smoking. Related video links are pushed and pulled to the users via Short Message Service (SMS) to support smoking cessation. Findings have implications for tobacco cessation treatment delivery and assessment of smoking status.

Duke Scholars

Author Ming-Chun Huang DKU Faculty