A New Mining Method to Detect Real Time Substance Use Events from Wearable Biosensor Data Stream

Area of Science:

• Biomedical Engineering
• Data Science
• Public Health

Background:

• Real-time substance use detection is crucial for effective behavioral interventions against drug abuse.
• Current methods like self-reporting and urine screening have significant limitations in timeliness and intrusiveness.
• Wearable biosensor technology presents a promising but underexplored avenue for real-time drug use monitoring.

Purpose of the Study:

• To develop and evaluate a novel method for real-time detection of substance use events using wearable biosensor data.
• To address the limitations of traditional drug detection methods for timely intervention.
• To establish thresholds for parameter detection in real-time substance use event identification.

Main Methods:

• Utilized a sliding window technique to process continuous data streams from wearable biosensors.
• Employed a distance-based outlier detection algorithm to identify potential substance use events.
• Performed numerical analyses to determine optimal parameter thresholds for event detection.

Main Results:

• Successfully developed a real-time substance use event detection method using wearable biosensor data.
• Empirically identified specific parameter thresholds for detecting cocaine use.
• Demonstrated the adaptability of the proposed method for detecting other substance use events.

Conclusions:

• The proposed method offers a viable solution for real-time substance use detection via wearable biosensors.
• This approach overcomes the limitations of traditional methods, enabling more timely interventions.
• The method shows potential for broad application in monitoring various substance use patterns.