Device and method for multi-dimensional classification of rodent vocalizations

Patent metadata

Field	Value
Publication number	US-10463017-B2
Application number	US-201715667862-A
Country	US
Kind code	B2
Filing date	Aug 3, 2017
Priority date	Aug 3, 2017
Publication date	Nov 5, 2019
Grant date	Nov 5, 2019

Abstract

The field of this invention is classifying animal behaviors. In particular the fields of this invention include using animals in vivariums, such as rodents, particularly mice. In a vivarium cage, animals create ultrasonic vocalizations. In addition, their phenotype changes over time. The vocalizations, a current phenotype, and the cage environments are inputs to a multi-dimensional classifier using clustering algorithms to find multi-dimensional close relationships. Such close relationships may be identified as particular behavior.

First claim

We claim:

1. A device for multi-dimensional classification of animal vocalizations comprising:
a study animal type, wherein the study animal type is non-human;
a vivarium comprising a set of study animals, of the study animal type, in a plurality of study cages;
a set of environmental attributes consistent with the plurality of study cages;
at least one ultrasonic audio sensor proximal to the each study cage of the plurality of study cages, adapted to receive, from the each study cage, audio vocalizations from at least one study animal of the set of study animals within the each of the study cage, and then communicate the audio vocalizations to a multi-dimensional processor;
at least one phenotype sensor proximal to each study cage of the plurality of study cages, adapted to detect and communicate one or more phenotype attributes of an animal in the each study cage;
the multi-dimensional processor adapted to accept as an input (i) the audio vocalizations, (ii) the phenotype attributes, and (iii) the set of environmental attributes, from the each of the at least one study animal;
wherein an output of the multi-dimensional processor comprises a set of multi-dimensional clusters responsive to the (i), (ii), and (iii) inputs;
wherein each multi-dimensional cluster in the set of multi-dimensional clusters is associated with an animal type, innate animal attribute, phenotype, or environmental attributes of the at least one study animal of the set of study animals within the each of the study cage.

2. The device of claim 1 , wherein:
the each study cage of the plurality of study cages is a home cage of a respective animal, and free of electronic penetrations.

3. The device of claim 1 , wherein:
the at least one ultrasonic acoustic sensor and the at least one phenotype sensor are mechanically independent of the each of the study cage such that the ultrasonic acoustic and the at least one phenotype sensors of the each of the study cage are replaceable without mechanically removing the each of the study cage or the ultrasonic and the at least one phenotype sensors, respectively; and
wherein the set of environmental attributes is predetermined.

4. The device of claim 1 , wherein:
the device is free of manually observed behavior of the set of study animals, and free of manually communicated behavior of the set of study animals, and free of manual classifying behaviors.

5. The device of claim 1 , wherein:
the communication of the audio vocalizations and the detection and communication of the phenotype attributes are continuous.

6. The device of claim 1 , wherein:
at least some study cages of the plurality of study cages comprise two or more animals of the set of study animals.

7. A method of classification of animal vocalizations using a device comprising:
a study animal type, wherein the study animal type is non-human;
a vivarium comprising a set of study animals, of the study animal type, in a plurality of study cages;
a set of environmental attributes consistent with the plurality of study cages;
at least one ultrasonic audio sensor proximal to the each study cage of the plurality of study cages, adapted to receive, from the each study cage, audio vocalizations from at least one study animal of the set of study animals within the each of the study cage, and then communicate the audio vocalizations to a multi-dimensional processor;
at least one phenotype sensor proximal to each study cage of the plurality of study cages, adapted to detect and communicate one or more phenotype attributes of an animal in the each study cage;
the multi-dimensional processor adapted to accept as an input (i) the audio vocalizations, (ii) the phenotype attributes, and (iii) the set of environmental attributes, from the each of the at least one study animal;
wherein an output of the multi-dimensional processor comprises a set of multi-dimensional clusters responsive to the (i), (ii), and (iii) inputs;
wherein each multi-dimensional cluster in the set of multi-dimensional clusters is associated with an animal type, innate animal attribute, phenotype, or environmental attributes of the at least one study animal of the set of study animals within the each of the study cage.

8. A system of classification of animal vocalizations using a device comprising:
a study animal type, wherein the study animal type is non-human;
a vivarium comprising a set of study animals, of the study animal type, in a plurality of study cages;
a set of environmental attributes consistent with the plurality of study cages;
at least one ultrasonic audio sensor proximal to the each study cage of the plurality of study cages, adapted to receive, from the each study cage, audio vocalizations from at least one study animal of the set of study animals within the each of the study cage, and then communicate the audio vocalizations to a multi-dimensional processor;
at least one phenotype sensor proximal to each study cage of the plurality of study cages, adapted to detect and communicate one or more phenotype attributes of an animal in the each study cage;
the multi-dimensional processor adapted to accept as an input (i) the audio vocalizations, (ii) the phenotype attributes, and (iii) the set of environmental attributes, from the each of the at least one study animal;
wherein an output of the multi-dimensional processor comprises a set of multi-dimensional clusters responsive to the (i), (ii), and (iii) inputs;
wherein each multi-dimensional cluster in the set of multi-dimensional clusters is associated with an animal type, innate animal attribute, phenotype, or environmental attributes of the at least one study animal of the set of study animals within the each of the study cage.

Assignees

• Vium Inc

Inventors

• Betts-Lacroix Jonathan
• Schaevitz Laura

Classifications

• G16H40/63

  for local operation · CPC title

• A61B5/1128

  using image analysis (A61B5/1127 takes precedence) · CPC title

• A61B2503/40

  Animals · CPC title

• G16H40/67

  for remote operation · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title