Behavioral analysis has heavily relied on manual annotation or other commercial platforms in the past, which are expensive, time-consuming, and vulnerable to variability. DeepLabCut (DLC) is an animal tracking program that utilizes pose estimation to generate data that would otherwise be challenging to collect. Although DLC accurately generates data, the lack of familiarity with specialized computer software has delayed the transition to this model. As a result, we utilized SimBa (Simple behavioral analysis), an open-source program that uses a graphical interface, to simplify the machine learning process. Through SimBa, we generated data on mice, such as distance covered and time occupied in one area in the context of complex social interactions. This preliminary data can be pivotal in explaining higher-order social interactions yet would have gone unnoticed without advanced pose estimation trackers in place.
More about DeepLabCut: http://www.mackenziemathislab.org/deeplabcut
More about SimBA: https://goldenneurolab.com/simba
Powered by Acadiate
© 2011-2024, Acadiate Inc. or its affiliates · Privacy