Although the advent of high-resolution imaging has enabled medical professionals and scientists to better understand the brain circuit dysfunction seen in epilepsy, much less is known about how epilepsy affects behavior. Is not … A new study uses state-of-the-art AI in mice to capture epilepsy-related behaviors that the human eye might miss.
Epilepsy is the most common chronic brain disease, affecting millions of people worldwide. It can affect people of all ages, and for some people, treatment not only causes nasty side effects, but it also fails to prevent seizures from occurring.
Traditional approaches to the diagnosis of epilepsy and assessment of treatment involve the use of continuous video electroencephalogram (EEG) monitoring over days or weeks. However, given the complexity and variability of the condition, and the fact that some seizures are not visible on the EEG, it can be a rather dull tool. Healthcare professionals must rely on their ability to view and analyze hours of video EEG recordings, often noticing subtle behavioral changes.
Now, researchers have used an AI technology called MoSeq (or motion sequencing) to analyze the behavior of epileptic mice and identify behavioral “fingerprints” that go unnoticed by the human eye.
MoSeq is a machine learning technology that trains unsupervised machines to identify repetitive patterns of behavior. After identifying behaviors, MoSeq provides a suite of visualization tools and statistical tests to help scientists understand those behaviors and compare them to different experimental conditions.
Using MoSeq to analyze 3D videos of freely moving mice allowed researchers to identify, track, and quantify mouse behavior. They found that the technique could better distinguish between epileptic and non-epileptic mice, outperforming trained human observers. Moreover, unlike conventional methods, a 1-hour video recording was required and no seizures were required prior to providing analysis.
Researchers were able to use AI to distinguish behavioral patterns in mice after being given one of three antiepileptic drugs.
The successful use of machine learning techniques will provide humans with faster, less labor-intensive, less costly, and more objective methods for diagnosing epilepsy and testing the efficacy of antiepileptic drugs. shows what is possible.
The study was published in a journal neuron.
Source: National Institute of Neurological Disorders and Stroke
