System and method of determining a condition of a subject based on volatile organic compounds

A method according to any one of the preceding claims, further comprising:

• receiving one or more initial signals from the one or more sensor;
• associating the initial signals as surrounding background signals;
• and
• filtering background noise from the set of sensor signals using the initial signals.

A system for determining a condition of a subject based on Volatile Organic Compounds (VOCs) in a gaseous phase, originating from the subject, the system comprising:

• one or more VOCs sensors configured to detect VOCs originated from the at least one subject;
• and
• a controller configured to:
• receive from the one or more sensors a set of sensor signals in response to exposing the one or more sensors to a sample of the VOCs;
• extract one or more feature values from the sensor signals;
• receive a classification model, trained to classify samples of VOCs based on the one or more extracted feature values correlated to one or more condition of the subject;
• associate the one or more extracted features received from the set of sensor signals with one or more classes of the classification model;
• and
• determine the condition of the subject based on the association.

A method of training a classification model to determine a condition of a subject, the method comprising:

• a. receiving from one or more sensors a set of sensor signals in response to exposing the one or more sensors to a sample of VOCs originated from the subject;
• b. extracting one or more feature values from the sensor signals;
• c. tagging the one or more feature values with a class associated with at least one known condition of the subject;
• and
• d. repeating steps (a) through (c) with a different sample, to train the classification model.

A system for training a classification model to determine a condition of a subject, the system comprising:

• one or more VOCs sensors configured to detect VOCs originated from the at least one subject;
• a storage unit;
• and
• a controller configured to:
• a. receive from one or more sensors a set of sensor signals in response to exposing the one or more sensors to a sample of VOCs originated from the subject;
• b. extract one or more feature values from the sensor signals;
• c. tag the one or more feature values with a class associated with at least one known condition of the subject;
• d. repeat steps (a) through (e) with a different sample, to train the classification model;
• and
• e. store the trained classification model in the storage unit.