Personal Automatic Cigarette Tracker (PACT)
This project aims at creation of a wearable device for monitoring of smoking and smoke exposure in free living individuals.
Cigarette smoking is the leading cause of preventable death in the United States. Smoking produces over 440,000 deaths each year in the US and generates an estimated $193 billion in annual health-related economic losses. Recent national surveys indicate that approximately 43.8 million people or 20% smoke. Worldwide, tobacco companies produce around 5.5 trillion cigarettes each year, enough for nearly 1000 cigarettes per human on the planet. World Health Organization estimates that annually 5 million people die worldwide from consequences of smoking, of which 500,000 dies from second hand smoke.
The American Cancer Society says about 70 percent of smokers want to quit, with about 40 percent attempting to quit each year. Smokers may try to quit five or six times before quitting for good. Some people can stop smoking on their own while some need assistance in the form of behavioral support programs and pharmacological interventions. The question remains why some people can and some cannot stop this habit on their own.
The health consequences of cigarette smoking are determined by the net exposure to cigarette smoke over time. Studies of smoking topography (which is comprised of variables such as maximum puff velocity, puff volume, and number of puffs) reveal that smokers vary considerably in the amount of smoke they inhale when they smoke a cigarette. Consequently, the number of cigarettes smoked over a given period is not strongly related to total smoke exposure for that same period. However, methods of smoking assessment available today do not permit the collection of accurate, non-reactive measures of smoking behavior that capture smoking frequency and comprehensive within-cigarette smoke exposure. The absence of accurate, real-time information about smoking behavior limits research on the natural dynamics of cigarette smoking and makes it difficult, for example, to comprehensively assess the impact of the use of various cigarette products on smoke exposure.
We are developing a Personal Automatic Cigarette Tracker (PACT). As envisioned, PACT does not rely on any form of self-report, does not interfere with the natural smoking behavior of an individual, and does not require any conscious effort to monitor smoking behavior in unconstrained settings. In addition, PACT can be used to assess measures of smoke exposure not available through conventional smoking topography devices. The data collected by PACT can also provide an objective method of assessing the effectiveness of behavioral and pharmacological smoking interventions. In addition, real-time smoking detection and smart phone feedback capabilities enable novel applications in monitoring of lapses and delivery of behavior-enhancing feedback during cessation interventions.
PACT relies on wearable sensors that monitor breathing and hand-to-mouth gestures of the individuals. The fact of cigarette smoking can be detected by computerized analysis of hand gesture and breathing signals and automatically detecting breathing patterns unique to smoke inhalations (puff, inhale, smoke holding and exhale). An early prototype of PACT is shown in the picture above, with the prototype under development being implemented as an elastic band worn on the chest level in a manner similar to the heart rate monitors. Details of PACT operation and results of validation testing can be found the journal and conference publications below.
- A Wearable Sensor System for Monitoring of Cigarette Smoking
- Detection of Hand-to-Mouth Gestures Using a RF Operated Proximity Sensor for Monitoring Cigarette Smoking
- Monitoring of cigarette smoking using wearable sensors and Support Vector Machines
Refereed Conference Publications
- Detection of cigarette smoke inhalations from respiratory signals using reduced feature set
- Comparative sensor analysis for an electronic wearable and non-invasive respiratory signal acquisition system
- Identification of Cigarette Smoke Inhalations from Wearable Sensor Data using a Support Vector Machine Classifier
- RF Hand Gesture Sensor for Monitoring of Cigarette Smoking
- Automatic breathing segmentation from wearable respiration sensors