Method and apparatus for measuring animal's condition by acquiring and analyzing its biological signals

• US 20030166996A1
• Filed: 01/10/2003
• Published: 09/04/2003
• Est. Priority Date: 01/11/2002
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Hyoung-ki, Kim, Kyung-hwan, Bang, Seok-won
• Granted Patent: US 7,282,028 B2
• Time in Patent Office: Days
• US Class Current: 600/300
• CPC Class Codes: A01K 29/00 Other apparatus for animal husbandry
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/16 Devices for psychotechnics using teaching or educational appliances G09B1/00 - G09B7/00; Testing reaction times ; Devices for evaluating the psychological state
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/24 Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics

Abstract

A method and apparatus for measuring the biological condition of an animal by acquiring and analyzing its biological signals are provided. The biological signals from skin temperature, a photoplenthysmogram (PPG), an electrocardiogram (ECG), electrodermal activity (EDA), an electromyogram (EMG), and an electrogastrogram (EGG) are detected using a biological signal detection unit which is attached to the animal'"'"'s skin. Feature vectors, including the mean heart rate of the photoplenthysmogram and its standard deviation, the very low frequency, low frequency, and high frequency components of heart rate variability, the frequency and mean amplitude of skin conductance responses, and the mean and maximum skin temperatures, are extracted from the detected biological signals. The biological condition, including needs and emotions, of the animal as to whether or not the animal feels hunger or fear, how much the animal is stressed, or whether or not the animal needs to have a bowel movement, is determined using a pattern classifier which has learned reference vectors, which reflect the behaviors, needs, and emotions of different kinds of animals for various biological conditions and are stored in a predetermined database. Therefore, the biological condition of the animal can be determined through instrumental communication, not through human languages, and the breeding of pets can be efficiently managed.

Claims

1. A method for determining the biological condition of an animal by acquiring and analyzing a biological signal, the method comprising:

(a) acquiring a predetermined biological signal using a biological signal detection unit which is attached to the animal;

(b) extracting a predetermined feature vector from the acquired predetermined biological signal; and

(c) analyzing and determining the biological condition of the animal, including emotions and needs of the animal, from the predetermined feature vector.

2. A method for determining the biological condition of an animal by acquiring and analyzing biological signals, the method comprising:

(a) acquiring biological signals from skin temperature, a photoplenthysmogram (PPG), an electrocardiogram (ECG), electrodermal activity (EDA), an electromyogram (EMG), and an electrogastrogram (EGG) using a biological signal detection unit which is attached to the animal'"'"'s skin;

(b) extracting feature vectors from the acquired biological signals, the feature vectors including the mean heart rate of the photoplenthysmogram and its standard deviation, the very low frequency, low frequency, and high frequency components of heart rate variability, the frequency and mean amplitude of skin conductance responses which are obtainable from the electroderminal activity, and the mean and maximum skin temperatures; and

(c) analyzing and determining the biological condition, including needs and emotions, of the animal as to whether or not the animal feels hunger or fear, how much the animal is stressed, or whether or not the animal needs to have a bowel movement, using a support vector machine which has learned reference vectors, which reflect the behaviors, needs, and emotions of different kinds of animals for various biological conditions and are stored in a predetermined database.

3. The method of claim 2, wherein extracting the mean heart rate as one of the feature vectors in step (b) from the photoplenthysmogram, which is acquired as a biological signal in step (a), comprises:

(b1) low-pass filtering the photoplenthysmogram to remove high frequency noise;

(b2) median-filtering an output signal in step (b1), from which the high frequency noise is removed, to estimate a baseline variation to a very low frequency;

(b3) subtracting an output signal in step (b2) from the output signal in step (b1) to obtain a waveform from which the baseline variation and direct current and high frequency components are removed; and

(b4) matching-filtering an output signal in step (b3) to obtain the location of peaks of the photoplenthysmogram and measure the heart rate and heart rate variability.

4. The method of claim 2, wherein extracting the very low frequency, low frequency, and high frequency components of heart rate variability as feature vectors in step (b) from the electrocardiogram, which is acquired as a biological signal in step (a), comprises:

(b1′

) bandpass-filtering the electrocardiogram to remove background noise which does not belong to the electrocardiographic frequency band;

(b2′

) median-filtering an output signal in step (b1′

) from which the background noise is removed to estimate a baseline variation;

(b3′

) subtracting an output signal in step (b1′

) from the output signal in step (b2′

) to obtain a waveform from which the baseline variation and noise which does not belong to the frequency band are removed;

(b4′

) applying an output signal in step (b3′

) to a Teager energy operator to detect R-peaks in which the amplitude and frequency abruptly rise;

(b5′

) measuring time-series heart rate variability from the intervals of the R-peaks; and

(b6′

) calculating the power of particular frequency bands from the measured time-series heart rate variability through autoregressive modelling using the Burg algorithm.

5. The method of claim 4, wherein when the standard deviation of heart rates is calculated as one of the feature vectors in step (b), the heart rates within a predetermined range of the maximum and minimum heart rates are excluded.

6. The method of claim 2, wherein extracting the frequency and amplitude of skin conductance responses as feature vectors in step (b) from the electrodermal activity, which is acquired as a biological signal in step (a), comprises:

(b1″

) down-sampling an input electrodermal activity waveform;

(b2″

) obtaining a differential waveform from an output waveform in step (b1″

);

(b3″

) smoothening the differential waveform in step (b2″

) by convolution with a Bartlett window having a predetermined length; and

(b4″

) determining two points of the smoothened waveform that have a predetermined threshold level as the start and end points of a skin conductance response, respectively.

7. The method of claim 2, wherein in step (c), whether or not the animal feels hunger is determined using a feature vector extracted from the electrogastogram acquired as a biological signal in step (a).

8. The method of claim 7, wherein in step (b), blind source separation is performed on the electrogastrogarm acquired in step (a) by independent component analysis in order to remove noise from an animal'"'"'s motion and an instrumental noise waveform, and the resulting electrograstrogram is cut to a time length and overlapped by a predetermined time period to obtain a time-varying spectrum through autoregressive modelling using the Burg algorithm as a feature vector indicating the degree of hunger of the animal, and

in step (c), based on a plurality of electrogastrograms measured in step (b), the degree of hunger is classified into a predetermined number of levels, 2-dimensional feature vector data which are constituted of spectral peaks and widths for each level are generated, and the degree of hunger of the animal is determined by comparing the 2-dimensional feature vector with the reference vector stored in a predetermined support vector machine.

9. The method of claim 2, wherein in step (c), whether or not the animal feels fear and how much the animal is stressed are determined using the feature vectors extracted from the electrocardiogram, electrodermal activity, and skin temperature acquired in step (a).

10. The method of claim 2, wherein in step (c), whether or not the animal needs to have a bowel movement is determined using a feature vector extracted from the electromyogram acquired in step (a) of anal muscles.

11. The method of claim 2, further comprising (d) informing an animal'"'"'s owner of the biological condition, including needs and emotions, of the animal via a predetermined communication unit.

12. The method of claim 2, further comprising:

(e) transmitting the biological signals of the animal acquired in step (a) and/or the biological condition, including needs and emotions, of the animal determined in step (c) to an animal'"'"'s handler via a predetermined communication unit; and

(f) the animal'"'"'s handler informing an animal'"'"'s owner of the result of a diagnosis performed using the biological signals and/or the biological condition of the animal via a predetermined communication unit.

13. An apparatus for measuring the biological condition of an animal by acquiring and analysing biological signals, the apparatus comprising:

a biological signal detection unit which is attached to the animal'"'"'s skin and detects a biological signal of the animal; and

a condition analysis unit which determines the biological condition, including needs and emotions, of the animal using a predetermined feature vector extracted by analyzing the biological signal detected by the biological signal detection unit.

14. The apparatus of claim 13, further comprising a transmission unit which wirelessly transmits the biological condition, including needs and emotions, of the animal that has been analyzed by the condition analysis unit to an animal'"'"'s owner.

15. The apparatus of claim 13, further comprising a transmission unit which wireless transmits the biological signal(s) detected by the biological signal detection unit or the biological condition determined by the condition analysis unit to a predetermined expert, including a veterinarian.

16. An apparatus for measuring the biological condition of an animal by acquiring and analysing biological signals, the apparatus comprising:

a biological signal detection unit including a plurality of sensors and electrodes, which is attached to the animal'"'"'s skin and detects biological signals from skin temperature, an electrocardiogram (ECG), a photoplenthysmogram (PPG), electrodermal activity (EDA), an electromyogram (EMG), and an electrogastrogram (EGG);

a feature vector extraction unit which extracts feature vectors from the biological signals detected by the biological signal detection unit, the feature vectors including the mean heart rate of the photoplenthysmogram and its standard deviation, the very low frequency, low frequency, and high frequency components of heart rate variability, the frequency and mean amplitude of skin conductance responses, and the mean and maximum skin temperatures;

a biological information database which stores reference vectors reflecting the behaviors, needs, and emotions of different kinds of animals for various biological conditions; and

a biological condition analysis unit which compares the feature vectors extracted by the feature vector extraction unit with reference vectors of the biological information database and determines the biological condition, including needs and emotions, of the animal as to whether or not the animal feels hunger or fear, how much the animal is stressed, or whether or not the animal needs to have a bowel movement.

17. The apparatus of claim 16, further comprising a transmission unit which wirelessly transmits the biological condition, including needs and emotions, of the animal that has been analyzed by the biological condition analysis unit to an animal'"'"'s owner.

18. The apparatus of claim 16, further comprising a transmission unit which wireless transmits the biological signal(s) detected by the biological signal detection unit or the biological condition determined by the biological condition analysis unit to a predetermined expert, including a veterinarian.