Mobile animal surveillance and distress monitoring
A method for remote animal surveillance and distress monitoring includes detecting biometric and behavioral parameters of the animal, identifying novel events based on comparison of detected parameters to predefined parameter values and qualifications; determining whether a composite parameter value exceeds a predefined composite threshold value indicative of possible distress in the animal; and notifying remote caretakers of possible distress in the animal based on the composite value exceeding the predefined composite threshold value.
1. A method for mobile point-of-care animal surveillance and distress monitoring comprising:
detecting one or more biometric parameter of the animal, wherein the one or more biometric parameter comprises one or more of a respiratory rate, heart rate, and temperature of the animal;
detecting one or more behavioral parameter of the animal, wherein detecting the one or more behavioral parameters comprises monitoring data from one or more of an accelerometer, gyroscope, magnetometer, and barometric pressure sensor;
preprocessing the one or more biometric or behavioral parameter to remove noise resulting from movement of the animal;
transforming the one or more detected biometric and behavioral parameter into the frequency domain to obtain one or more frequency coefficient;
multiplying the one or more frequency coefficient across a weighted matrix to generate a matrix characterizing components of the one or more detected biometric and behavioral parameter;
determining, using a one-class classifier, occurrence of a novel event based on comparison of the components of the one or more detected biometric and behavioral parameter to a range of predefined personalized historical parameter values;
notifying one or more remote caretakers of possible distress in the animal based on one or more of the components of the one or more detected biometric and behavioral parameter exceeding the range of predefined personalized historical parameter values.
2. The method of claim 1 , wherein two or more of the detected biometric and behavioral parameters are combined into a composite value before transformation into the one or more frequency coefficient.
3. The method of claim 1 , wherein a Gaussian Mixture Model represents one or more detected biometric and behavioral parameter values under normal conditions.
4. The method of claim 1 , further comprising detecting one or more biologic function parameter of the animal and using the detected biologic function parameters for at least one of determining the occurrence of a novel event and computing the one or more frequency coefficient.
5. The method of claim 2 , further comprising continuously personalizing the range of predefined parameter values to conform to the one or more detected biometric and behavioral parameters specific to the animal over time.
6. The method of claim 1 , further comprising use of fuzzy logic to aggregate the one or more biometric, biologic, and behavioral parameters to derive a single metric proportional to an animal level of distress, wherein the one or more aggregated biometric, biologic, and behavioral parameters includes at least one of a length of time of a novel event, deviation of a novel event from normality, and a count of surrounding novel events.
7. The method of claim 1 , wherein notifying one or more caretakers comprises activation of an escalating notification protocol across multiple channels.