• Here, we designed a real-time 3D motion reconstruction system to detect whole freely moving behavior in open-field with marker-less.
• The AVATAR system consist of multi-vision hardware and object detection deep learning algorithm that can be used for various mice experiment.
• This system precisely identifies motion units, a set of body posture/frame, enabling the real-time motion sequencing with sub-millisecond resolution, opening new avenue toward the unsupervised behavioral analysis.
• Also, we provide the web visualization service for AVATAR behavior data set.

Product Description

A next-generation system for quantification of animal behavior experiments that can replace the existing preclinical scoring that consists of human standards. As a fusion system of multi-vision hardware and AI software, all behaviors of mice, a representative animal of preclinical testing, can be quantified in real time with markerless and simulated in a 3D virtual space.

1. Engineering: Describe the engineering and functionality of the product and how they relate to the product’s intended use. Note: This may include materials, components, distinctive features, details about user value, processes used to fabricate the product or proprietary algorithms.

This system is a fusion of specially designed multi-vision hardware and deep learning software. It can produce real-time analysis of the external expression behavior of the mouse, which is most used in preclinical experiments, with quantifying 8 major skeleton vectors. For this purpose, hardware part was specially developed with observation chamber for mouse model and 5 cameras. And our software detects the main body-points in real-time from the images captured by the each camera from multiple angles, and can reconstruct and simulate them in 3D space. Researchers can analyze the overall behavior of mice by just placing them in a chamber without any pretreatment and marker.

2. Design: Describe the product's aesthetic and design qualities. Note: This may include functions and how they relate to user benefit, materials, components, ecological impact, platform(s) on which you intend to run, and/or channel(s) through which it is distributed.

The design of this product takes many factors into consideration for automation and refinement of pre-clinical testing. First, in order to minimize the influence of external stimuli, the entire equipment was made of an opaque acrylic cover, and a transparent observation chamber was detachable in it, so that the imaging quality of behavior of mice could be preserved in anyplace. Second, the chamber itself was made in a spherical shape to prevent the change of the visual stimulus according to the shape of the external chamber that was changed according to the direction of the rat. The software platform is web-based for easy access by researchers, and the UI is focused on 3D skeleton simulation for observation of human researcher.

3. Innovation: List up to three key points about the product and what makes it unique. Note: This may include innovative features, life-changing technology, patents or licensing of the technology and/or other reasons why it deserves an award.

Animals move according to commands from the brain. Careful quantification of the motion allows for reasoning about brain functions. To quantify the motion of a mouse in a laboratory, those who conduct the experiment have to continue to observe the mouse visually, play the recorded videos repeatedly, and record and analyze the results. Visual analysis of the motion involves subjective viewpoints and errors. This product was developed to automatically observe and quantify the motion in animal experiments. it enabled the researchers to increase the accuracy of the animal motion experiments and find new disease motions that had not been revealed before.