I am interested in your experience in automated (semi-automated) scoring systems of behavioural tests like: locomotion, FST, social interactions, NOR, Y maze, EPM, NSFT, etc.
Also, are there any useful freewares? Or is it necessary to buy EthoVision or something similar?
Thanks a lot!
Jana
Hello,
I've used EthoVision from Noldus and it's very practical to use, the new version has a lot of automatized functions and with the behavior recognition module it is even easier to analyze Social Intereaction and similar tests. I think it's worth it to invest the money on it.
Hi, in the past I've sued SMART and Any Maze. Both were really easy to use and dont cost the earth. On caveat is automated scoring isn't as reliable as manual, whichever software you choose.
I personally advice you score manually. From personally experience with scoring hundreds of hours of rodent behaviour on the Elevated Plus Maze, rodents can exhibit some strange behaviour that may not be reliably picked up or characterised using automation.
I have been using Anymaze software for tracking mice and rat for three years and it is really a fantastic software for EPM, NOR, FST, fear conditioning,open field, y maze,social interaction,water maze. It is more accurate than manual scoring for freezing , immobility, object exploitation within two centimeters,mobility,time in open arm. Apparatus and Protocol set up is very very important. Once you set up everything, just put in your animal,then you will have the result. I can share my protocol if you want.
Thank you all for the useful advice! It seems there are two religions, the manual and the automated :-) I have met many people (companies, mostly) claiming anything manual cannot be standard and therefore reliable. We will hopefully be able to buy an automated system and ask more questions then.
Jana Ruda-Kucerova
We offer a secure cloud-based app that allows you to automate your animal behavioral research. And you can manually score too!
It can be accessed anywhere in the world by a computer, smartphone, or tablet without downloading anything.
Wouldn't it be amazing to do your behavioral analysis in front of the TV watching Netflix...
You don't need anything fancy in order to get started, typically you can set up your smartphone, and let it record, while you're in a separate room. There is no fear of an interruptance caused by a researcher, etc.
Although we are not an open-source program. Our prices are very competitive when compared to Noldus' EthoVision. You can even share your data with your collaborators.
Please feel free to start your free trial at www.behaviorcloud.com or email me at [email protected] for more details.
I think the question is wrong. To do automated behavioral scoring you are not restricted to video. There are better sensor technologies than video to automatically classify behaviors. An independent study of MIT from about 8 years ago showed that capturing animal behavior requires 4 high speed high-resolution camera's. This is only the "capturing" of the content. After this you still need the software (classification algoritms) to reliably identify each behavior. Video data is still big and therefore analysis with complicated algorithms is taking significant time, limiting the possibilities of the methods that can be applied.
Behavior can basically be characterized as a combination of movements involving specific body parts or the entire body of the animal. From a physics point of view, this is basically kinetic energy that you are looking at.
Consequently, video is a very indirect way of capturing behavior, because it is nothing more than a stream of photos showing different postures of the animal. You can also understand this better by slowing down the video speed. At some point the video uses it relevance and it becomes unclear what the animal is foing.
Video is not measuring force or energy, which is a critical parameter in describing behaviors properly from both an ethological and physics point of view.
The reason that many researchers use video has to do with the fact that traditionally behavior scoring is based on visual observation and most automation was limited for a long time to video tracking (capturing only the locomotion of the animal). Visual manual observation is a method that is proven to be unreliable without proper training (expressed by large interobserver deviations). The fact is that training in reliality is nothing more than being able in predicting better what the animal is going to do. The question in such an approach is however if your score is objective when it is based on prediction based on prior experience and knowledge. In phenotyping and pharmacological experiments behavioral sequences might be altered.
Moreover, it is our experience that many researchers when asked to describe a specific behavior in detail are not able to give a description that uniquely describes the behavior (which is also not easy to do). As a result it is impossible that two observers using such a description will get exactly the same result
In summary, it's all about making proper definitions of behaviors first and then choosing the best measurement method in terms of really measuring the behavior and then developing the right technologies (algorithms) to automatically analyze and classify the behaviors of laboratory animals.
Kind Regards,
Ronald Bulthuis
(Managing Director Metris b.v.)
www.metris.nl