Brown University

Vium, Inc.

Youssef Barhomi

Jonathan Betts-Lacroix

Kyle Howard Heath

Daniel J Ford

Thomas Serre

Kevin Bath

Pablo Jadzinsky

Sven Eberhardt

Zachary Nado

The field of this invention is recording motion of an animal, uniquely identifying the animal, and recording an activity of that animal. Motions of animals on a path when the animal is or can be uniquely identified are tracklets. Tracklets begin and end at ambiguation events, where these are defined as locations and times of an animal where it cannot be uniquely identified. A first animal may be uniquely identified by first identifying all other animals in the first animal's environment. Animal identification may be after the end of a tracklet. Embodiments use optical flow analysis from video of an animal's environment. Embodiments record an animal's activity on tracklets and then use that activity to measure animal health or use that activity as data for a study using animals.

Tracklets

The method and system includes a plurality of test subject containers, each having a test subject therein. A plurality of video cameras is focused on the test subject containers to capture video of the behavior of each of the test subjects. The system and method also includes a storage system for storing the plurality of video from the cameras. The storage system may be local or cloud-based as is known in the art. The system further has one or more computers with a neural network configured to (a) retrieve the video from the storage system of the test subjects, (b) analyze the video to identify a plurality of observable behaviors in the test subjects, (c) annotate the video with the observed behavior classifications, and (d) store the annotated video in the storage system.

Method and system for automated behavior classification of test subjects

Growing community of inventors

Arlington, MA, United States of America

Youssef Barhomi