System and method for detection of cravings in individuals with addiction

A method for detecting drug craving of an individual subject, comprising:

• the individual subject undergoing a period of supervision wearing physiological sensors, the physiological sensors configured to measure training motion of the individual subject in three dimensions, training Electrodermal Response (EDR), and training temperature;
• recording windows of training motion data in the three dimension, training EDR data, and training temperature data by sensing the training motion, the training EDR, and the training temperature of the subject during a period of no drug use;
• generating training mean and variance of:
• the windows of the training motion data;
• training EDR data;
• and training temperature data, during the period of no drug use;
• applying machine learning to determining classification rules during the period of no drug use;
• releasing the individual subject from the period of supervision to a period of monitoring;
• recording:
• monitoring motion data in the three dimensions;
• monitoring EDR data;
• and monitoring temperature data, during the period of monitoring;
• generating monitoring mean and variance of:
• the monitoring motion data;
• monitoring EDR data;
• and monitoring temperature data, during the period of monitoring;
• comparing the monitoring mean and variance during the period of monitoring to the classification rules;
• and
• providing a warning of a condition selected from the group consisting of stress and drug cravings when the comparison indicates the presence of the condition.

A method for detecting drug craving of an individual subject, comprising:

• the individual subject undergoing a period of supervision wearing physiological sensors, the physiological sensors configured to measure training motion of the individual subject in three dimensions, training Electrodermal Response (EDR), and training temperature;
• recording training windows of training motion data in the three dimension, training EDR data, and training temperature data by sensing the training motion, the training EDR, and the training temperature of the individual subject during a period of no drug use;
• generating training mean and variance of the training windows of the training motion data, training EDR data, and training temperature data during the period of no drug use;
• transforming the three dimensional training motion data in the training windows into training amplitude data;
• creating training histogram data of the training amplitude data;
• fitting a training curve to the training histogram data;
• determining training shape and scale from the training curve;
• applying a machine learning to the training mean and variance and training shape and scale determining classification rules during the period of no drug use;
• recording monitoring windows of monitoring motion data in the three dimension, monitoring EDR data, and monitoring temperature data by sensing the monitoring motion, the monitoring EDR, and the monitoring temperature of the individual subject during a period of no drug use;
• generating monitoring mean and variance of the monitoring windows of the monitoring motion data, monitoring EDR data, and monitoring temperature data during the period of no drug use;
• transforming the three dimensional monitoring motion data in the monitoring windows into monitoring amplitude data;
• creating monitoring histogram data of the monitoring amplitude data;
• fitting a monitoring curve to the monitoring histogram data;
• determining monitoring shape and scale from the monitoring curve;
• comparing the monitoring mean and variance and monitored shape and scale during the period of monitoring to the classification rules;
• and
• providing a warning of a condition selected from the group consisting of stress, drug cravings, and drug use when the comparison indicates the presence of the condition.