Knowledge Base

System Configurations

EthoVision® XT Base allows you to perform complete tests with one animal in one arena. You can extend the functionality with one or more of the eight available modules:

• Mouse Behavior Recognition Module – Automatically detect ten mouse behaviors such as grooming, rearing, and digging.
• Rat Behavior Recognition Module – Automatically detect ten rat behaviors such as grooming, rearing, and sniffing.
• External Data Module – Integrate EthoVision XT data with co-acquired external data.
• Multiple Arenas Module – Track animals in up to a hundred arenas simultaneously.
• Social Interaction Module – Track multiple animals within each arena.
• Trial & Hardware Control Module – Create complete test protocols to initiate and control interaction with external equipment, based on real-time analysis of the animal's behavior.
• Quality Assurance Module – Assign different rights to different users and comply with Good Laboratory Practice guidelines.

Data Management

1. Setup

Template Experiments

EthoVision XT contains a user-guided set-up with templates for several test paradigms. At the start of the program you can either start a brand-new experiment, re-use the settings from a previous experiment, or use the user-guided set-up to open an experiment template suited to a specific test (water maze, plus maze, etc.). With these templates, many basic settings are pre-set, such as arena definition, detection settings, and parameters for analysis. You can fine-tune these settings to fit your specific needs.

DanioVision Experiments

If you use the DanioVision system for multi-well plates you can select the DanioVision experiment option which optimizes the arena settings and detection settings for zebrafish larvae in well plates.

Live Mouse Tracker Experiments

If you use the Live Mouse Tracker acquisition system and want to analyze these data in EthoVision XT you can select the Live Mouse Tracker experiment option when creating a new experiment. This creates a modified version of a standard, multi-subject, multi-body point experiment, with predefined, permanent settings for Live Mouse Tracker.

Good Laboratory Practice

For those conducting Quality Assurance experiments, particularly in environments adhering to Good Laboratory Practice (GLP) standards, EthoVision XT offers specialized options to meet these rigorous requirements. This ensures that your experiments not only yield reliable data but also comply with industry standards for quality and reproducibility.

Flexibility

For researchers seeking maximum flexibility in their experimental design, the "New from Scratch" option allows for complete customization of the experiment setup. This feature is ideal for innovative research that goes beyond standard protocols, offering the freedom to tailor every aspect of the experiment to unique research needs.

Experiment Options

• Tracking can be done live or from a video file.
• You can use digital cameras or analog cameras (but not a combination of the two).
• You can connect external data acquisition systems to EthoVision XT.
• Track in one or multiple arenas, with a maximum of 100 arenas.
• Track one or multiple subjects per arena, with a maximum of 16 subject roles.
• For mice, rats, and zebrafish you can use center point tracking, or center of gravity, nose, and tail base. For mice and rats you have the option between a contour-based technique or a deep learning based technique. You can also track based on color or color marker.
• Activity analysis – a non-tracking option analyzing overall activity per arena based on pixel changes. This is very useful for freezing detection.
• Distance can be measured in millimeters, centimeters, meters, or inches.
• Time can be measured in milliseconds, seconds, minutes, or hours.
• Rotation can be measured in degrees, radians, gradians, or rotations.

Manual Scoring

The Manual Event Recorder is a built-in coding system which allows you to define a list of behaviors that you can score manually during and after data acquisition (live or offline).

• Score live.
• Score (also) offline, to score at a lower or faster playback speed. You can also edit scored events.

Types of behaviors:

• Start-Stop Behaviors – behaviors that are independent from each other.
• Point Events – behaviors without a duration.
• Mutually Exclusive Behaviors – behaviors grouped in such a way that activating one behavior automatically stops the other behavior.

You can define different behaviors and score those categories with keyboard keys or with the mouse. Without key codes defined, you can manually score events using the mouse.

Arena Settings

The arena is the region in the video image where the animal is tracked. You can define:

• The arena(s).
• Subdivisions of the arena (zones).
• Cumulative zones (define individual smaller zones to be taken into account for cumulative analysis).
• Hidden zones (used for shelters) and entry zones to analyze entries and exits.
• Zones organized in layers, so you can define and analyze zones that overlap.
• Points of interest.

You can define an unlimited number of zones in an arena and create zones of any shape. Calibration of your arena will allow EthoVision XT to calculate parameters such as velocity and distances to certain points. If you are tracking in multiple arenas, you can copy and paste the settings from one arena to the other.

Multiple Arena Definition

When your experiment contains multiple arenas – for instance a group of open fields or home cages – you can use the multiple arena set-up functionality. This means that you do not need to draw each arena separately. EthoVision XT allows you to easily rotate, scale, transpose, resize, and reposition the series of arenas simultaneously. In DanioVision experiments, automated arena detection detects the individual wells in standardized well-plates automatically.

Arena and Zone Auto-Subdivision

This option allows the automated subdivision of circular and rectangular arenas into multiple equal zones, including auto-labeling of these subdivisions. For example, it can be used to divide a water maze into quadrants, simplifying the process of analyzing behavior across different sections of the arena.

Multi-Camera Set-Up

In EthoVision XT you can combine the input of multiple cameras. This way you can track from multiple arenas (in combination with the Multiple Arenas Module) without the need to place them all under one camera.

Trial Control Options

You can program EthoVision XT to automatically start and stop tracking, based on any of the parameters offered by EthoVision XT. This feature also includes an external command option, which allows you to start another computer program when a condition becomes true. For example, present a visual stimulus on a computer screen or play a sound file when the animal enters a particular zone.

With the additional Trial & Hardware Control Module, you can define complete test protocols to initiate and control interactions between your animals and external sensor and stimuli devices. Real-time analysis of any of the parameters of EthoVision XT allows you to control complete test protocols based on your animal's behavior. You can use it for complex learning tasks such as the radial arm maze with automatically controlled doors and food delivery using pellet dispensers. All standard TTL-based hardware can be attached and controlled.

If you are using multiple arenas, conditions are applied to each arena individually.

Tracking Methods for One Animal Per Arena (Body Point Estimate)

The tracking method specifies how the body points are estimated once the subject is detected. Depending on how many subjects and body points you specify in the Experiment Settings, one or more methods for tracking of body points are available in the Method section of the Advanced Detection Settings.

Deep Learning Based Body Point Detection Technique

When using this technique, a new algorithm based on a convolutional neural network finds the nose and tail-base points of rodents with high accuracy and few or no nose-tail swaps. The network was trained to extract features from video images of rodents of various colors and in various backgrounds, with videos in which the nose and the tail-base were annotated. During tracking, the network analyzes a portion of the image that includes the detected subject. It creates a map of probability of occurrence for both the nose and tail-base points, and finally makes an estimate of the position of the nose point and the tail-base point based on the highest probability.

Contour-Based Body Point Detection Technique

• Any Species – Default – This detection method analyzes the contour of the area detected as subject at each sample to assign the nose point and tail base. Make sure in the detection settings that the tail is fully detected. With this method it may be possible to track non-rodent shapes but the method is not designed for it. This method is automatically activated when you track the center point in one animal per arena and you specified Other Species in the automated setup.
• Rodents – Default – This detection method analyzes the varying shape of the contour of the area detected as subject and builds up a rodent model. It is more robust than the shape-based method because it does not require the nose and tail to be visible: it can predict the position of the nose and the tail based on previous samples. Make sure in the detection settings that the tail is removed from the body contour with Erode and Dilate. Choose this method when you want to track a single rodent without occlusions or without difficult tracking conditions.
• Rodents – For Occlusions – This method can handle severe shape distortions, for example when the animal's body is occluded or when multiple animals touch. However, it requires a lot of computer performance. Choose this method for rodents in a social interaction context, or when you track rodents that can be occluded, for example by bars or other objects in the cage. Make sure in the detection settings that the tail is removed from the body contour with Erode and Dilate.
• Other Species – For Occlusions – This method does not make any assumption about the shape of the subject. Choose this method to track, for example, insects, crustaceans, and large animals. Also use this method if you track fish from the side view of the water tank and the Adult Fish method does not give good results. However, with this method you only track the center point.

Contour adjustments can be made in one of two ways or by combining both methods:

• Erosion – decreases the animal's surface area by 1–10 pixels.
• Dilation – increases the animal's surface area by 1–10 pixels.

Besides the above-mentioned options to optimize the detection and identification of subjects, contrast settings can be fine-tuned, subject size can be defined, and video smoothing can be applied. There are also some additional detection settings specific for when you are using the Rat or Mouse Behavior Recognition Module.

Tracking Methods for Multiple Subjects Per Arena

Color marker tracking uses the color marker itself to track the subject and identify it.

Contour-based body point detection offers two options:

1. Center of gravity tracking only.
2. Center of gravity, nose point, and tail-base tracking.

In this case two subject identification methods are available:

• Marker-Assisted Identification – Identities are distinguished based on a color marker.
• Unmarked Subjects – Identities may switch.

Source: Technical Specs - EVXT 18 - Rodents, Noldus Information Technology