Researchers at the University of California, San Diego have made significant progress in using artificial intelligence to follow animal and human behaviour with previously unheard-of accuracy and adaptability. Their ground-breaking invention, "GlowTrack," has the potential to revolutionise the way movement is studied by scientists and provide fresh perspectives on how the brain functions and governs the body.
Scientists have been tracking and analysing movement for years using a variety of methodologies, from old-fashioned clipboard-and-pen observations to more recent AI-based ones. These methods have developed over time, but they frequently rely on time-consuming, manual procedures that necessitate researchers to meticulously mark body parts hundreds or even thousands of times in order to properly train AI models.
GlowTrack adopts a completely new strategy, led by Associate Professor Eiman Azim and his group. Using fluorescent dye markers, this non-invasive movement tracking technique provides an effective, high-definition solution that can track even the smallest details.
Their innovative method, published in a paper in Nature Communications, opens up a plethora of opportunities in a variety of fields, including biology, robotics, medicine, and more. With the introduction of strong AI technologies into the lab, behaviour tracking has undergone a revolution.
There are disadvantages to using traditional methods to record animal movements. They depend on labour-intensive, manual procedures that are prone to limitations and human mistakes. Because AI models become experts at the training data they are given, human annotation restricts the application of these techniques to controlled testing contexts. Changes in lighting, orientation, camera angles, or other factors can cause the model to lose its ability to identify the tracked body parts.
The researchers labelled different bodily sections on humans or animals using fluorescent dye markers in order to get around these restrictions. Without the need for human annotation, these "invisible" markers quickly produce a large and visually diverse dataset that may be utilised to train AI models. The models can follow motions across a wide range of situations once fed this rich dataset; manual labelling would almost certainly result in resolutions that are not achievable.
The utilisation of fluorescent dye markers has revolutionised the field by simplifying the comparison of movement data among various research. The same models can be used by laboratories all over the world to monitor body motions in a variety of contexts, improving experiment comparability and reproducibility—two vital components of scientific research. GlowTrack has a bright future ahead of it since it combines the capabilities of various three-dimensional movement reconstruction technologies. Furthermore, these enormous movement records can be subjected to sophisticated analysis techniques to reveal hidden patterns.
The UC San Diego CMG Training Programme, a Jesse and Caryl Philips Foundation Award, the National Institutes of Health, the Searle Scholars Programme, the Pew Charitable Trusts, and the McKnight Foundation all provided funding for this ground-breaking research.
In conclusion, GlowTrack's introduction is poised to completely transform the field of animal and human behaviour research by providing a revolutionary, highly accurate, and adaptable approach that makes use of fluorescent dye markers and artificial intelligence. This innovative technology has the potential to revolutionise scientific research, improve our comprehension of movement, and offer insightful information about neurological conditions.