Event Title
Machine Learning and Reaching Behavior Patterns in Awake Mouse Model
Location
Suwanee, GA
Start Date
3-5-2022 1:00 PM
End Date
3-5-2022 4:00 PM
Description
The cerebellum plays a crucial role in coordinating motor control. Attaining more information on the association between cerebellar physiology and behavioral patterns in mice will allow for a better understanding of cerebellar disorders. One of the several forms of examinations conducted to evaluate cerebellar dysfunction in patients is the finger-nose-finger test, which is especially useful in evaluating dysmetria.
This experiment utilizes the machine learning software Deep Lab Cut (DLC) to track the reaching behavior. The goal is to correlate frame data from the DLC analysis with the firing of an isolated cerebellar neuron from an awake mouse to observe the firing patterns associated with each step of the reaching behavior. In the preliminary experimental design, a mouse under water restriction with limited head movement reached for a water droplet while being recorded from 2 angles. The recordings were then analyzed using DLC and correlated with reaching behavior. 3 mice were handled and one mouse was put under water restriction for this trial.
Analysis of these recordings shows that the mice were accurately tracked showing reaching by hand. The trajectory plots obtained sets a baseline for the reaching behavior frequency expected in mice who undergo water restriction. From the lateral view, the mouse’s reaching pattern is a straight line to and from the droplet with the longest duration spent at the droplet. From the inferior view the reaching pattern is curved to and from the droplet. Future experiments will utilize this data to refine the reaching behavior, including evaluating the reaching patterns shown in the non-ataxic mice and ataxic mice and their cerebellar activity.
Embargo Period
5-26-2022
Machine Learning and Reaching Behavior Patterns in Awake Mouse Model
Suwanee, GA
The cerebellum plays a crucial role in coordinating motor control. Attaining more information on the association between cerebellar physiology and behavioral patterns in mice will allow for a better understanding of cerebellar disorders. One of the several forms of examinations conducted to evaluate cerebellar dysfunction in patients is the finger-nose-finger test, which is especially useful in evaluating dysmetria.
This experiment utilizes the machine learning software Deep Lab Cut (DLC) to track the reaching behavior. The goal is to correlate frame data from the DLC analysis with the firing of an isolated cerebellar neuron from an awake mouse to observe the firing patterns associated with each step of the reaching behavior. In the preliminary experimental design, a mouse under water restriction with limited head movement reached for a water droplet while being recorded from 2 angles. The recordings were then analyzed using DLC and correlated with reaching behavior. 3 mice were handled and one mouse was put under water restriction for this trial.
Analysis of these recordings shows that the mice were accurately tracked showing reaching by hand. The trajectory plots obtained sets a baseline for the reaching behavior frequency expected in mice who undergo water restriction. From the lateral view, the mouse’s reaching pattern is a straight line to and from the droplet with the longest duration spent at the droplet. From the inferior view the reaching pattern is curved to and from the droplet. Future experiments will utilize this data to refine the reaching behavior, including evaluating the reaching patterns shown in the non-ataxic mice and ataxic mice and their cerebellar activity.