Volatile organic compound detection and classification

A computer program product for volatile organic compounds classification, the computer program product comprising one or more computer readable storage devices and collectively stored program instructions on the one or more computer readable storage devices, the stored program instructions comprising:

• program instructions to subject a set of sensors of an electronic nose to a temperature oscillation;
• program instructions to gather electronic nose voltage outputs for a set of a known volatile organic compounds (VOCs) as the set of sensors of the electronic nose undergo the temperature oscillation;
• program instructions to learn a set of data features corresponding to each of the known VOCs;
• program instructions to receive test data associated with detecting VOCs;
• program instructions to analyze the test data according to the set of data features associated with the known VOCs;
• program instructions to determine a match between each feature of the test data and a corresponding feature of the set of data features, yielding a set of matches, wherein determining a match between each feature of the test data and the corresponding feature of a set of data features comprises determining a match according to a second degree of anomaly;
• program instructions to define a first degree of anomaly for the test data according to the set of matches;
• and
• program instructions to classify the test data according to the first degree of anomaly.

A computer-implemented method for volatile organic compounds classification according to an array of sensors whose voltage output is a function of volatile organic compounds (VOCs) present in a given test sample that is in contact with the array, the method comprising:

• subjecting, by one or more processors, a set of sensors of an electronic nose to a temperature oscillation;
• gathering, by one or more computer processors, sensor voltage outputs for a set of known volatile organic compounds (VOCs) as the sensors undergo the temperature oscillation;
• learning, by the one or more computer processors, a set of data features corresponding to each of the known VOCs according to the voltage outputs;
• populating a database according to the sets of features;
• receiving a sample voltage output from the array, the voltage output associated with a test sample;
• analyzing, by one or more computer processors, the sample voltage output to match the features that correspond to various organic compounds;
• for each feature associated with each VOC of a database of known VOCs, determining whether the test sample voltage output matches the feature according to a first threshold for the feature;
• determining a first degree of anomaly for the test sample, the first degree of anomaly comprising the number of unmatched features;
• and
• classifying the test sample according to the sample voltage output matches.