System and a method of lab animal observation
A system and method for automating a recorded observed behavior of a laboratory animal in a clinical trial is disclosed. The present invention records a video of the laboratory animal with respect to one or more objects. The recorded data is utilized to detect at least one feature points such as head, whiskers, tail, waist etc. of the animal and a one or more objects in video, wherein the one or more objects are a known object and an unknown object. The present invention thus automatically determines the behavior characteristics of the animal by measuring a proximity and comparing the proximity with a predefined threshold criterion with respect to the one or more objects without human intervention. The behavior characteristics are determined in the form of a type of behavioral events, a time interval, number of occurrences of each type of behavioral events, latency, and distance travelled by the animal.
1. A method for automatically determining behavior characteristics of an animal during clinical trials, the method comprising:
detecting, by a detection module, at least one feature points of said animal and one or more objects in at least one video image, wherein said one or more objects comprises at least one known object and at least one unknown object;
measuring, by a measurement module, a proximity between said at least one feature points of said animal and said one or more objects;
automatically determining behavior characteristics, by a behavior determination module, of said animal by detecting that said proximity meets a predefined threshold criteria, wherein said behavior characteristics comprises at least one of a type of behavioral events of said animal, a time interval corresponding to each type of behavioral events, number of occurrences of each type of behavioral events, latency associated with each type of behavioral events, and distance travelled by said animal;
providing a notification message, by a notification module, indicating said behavior characteristics of said animal;
the type of behavioral events comprises sniffing at one or more objects and poking at one or more objects; and
the sniffing is an event counted when proximity between said at least one feature points and said at least one known object or said at least one unknown object meets a first pre-defined threshold criteria, wherein said first pre-defined threshold criteria is associated with a first value.
2. A method for automatically determining behavior characteristics of an animal during clinical trials, the method comprising:
detecting, by a detection module, at least one feature points of said animal and one or more objects in at least one video image, wherein said one or more objects comprises at least one known object and at least one unknown object;
measuring, by a measurement module, a proximity between said at least one feature points of said animal and said one or more objects;
automatically determining behavior characteristics, by a behavior determination module, of said animal by detecting that said proximity meets a predefined threshold criteria, wherein said behavior characteristics comprises at least one of a type of behavioral events of said animal, a time interval corresponding to each type of behavioral events, number of occurrences of each type of behavioral events, latency associated with each type of behavioral events, and distance travelled by said animal;
providing a notification message, by a notification module, indicating said behavior characteristics of said animal;
the type of behavioral events comprises sniffing at one or more objects and poking at one or more objects; and
the poking is an event counted when proximity between said at least one feature points and said at least one known object or said at least one unknown object meets a second pre-defined threshold criteria, wherein said second pre-defined threshold criteria is associated with a second value.
3. The method of claim 2 , wherein said second value is greater than said first value.
4. A system for automatically determining behavior characteristics of an animal, the system comprising:
at least one processor; and
a memory having stored instructions, the instructions causing the system to perform functions when executed by the at least one processor, the functions comprising:
detecting, by a detection module, at least one feature points of an animal and one or more objects in at least one video image, wherein said one or more objects comprises at least one known object and at least one unknown object;
measuring, by a measurement module, a proximity between said at least one feature points of said animal and one or more objects;
automatically determining behavior characteristics, by a behavior determination module, of said animal by detecting that said proximity meets a predefined threshold criteria, wherein said behavior characteristics comprises at least one of a type of behavioral events of said animal, a time interval corresponding to each type of behavioral events, number of occurrences of each type of behavioral events, latency associated with each type of behavioral events, and distance travelled by said animal;
providing a notification message, by a notification module, indicating said behavior characteristics of said animal;
the type of behavioral events comprises sniffing at one or more objects and poking at one or more objects; and
the sniffing is an event counted when proximity between said at least one feature points and said at least one known object or said at least one unknown object meets a first pre-defined threshold criteria, wherein said first pre-defined threshold criteria is associated with a first value.
5. The system of claim 4 , wherein said poking is an event counted when proximity between said at least one feature points and said at least one known object or said at least one unknown object meets a second pre-defined threshold criteria, wherein said second pre-defined threshold criteria is associated with a second value.
6. The system of claim 5 , wherein said second value is greater than said first value.