A team of researchers at Stanford has found a new way to potentially help patients with epilepsy. The researchers studied mice with acquired and genetic epilepsies, and they found that machine analysis can work better to identify epileptic vs. non-epileptic mice compared to trained human observers, according to New Atlas.
Motion Sequencing (MoSeq)
MoSeq is an AI technology the researchers used for behavior analysis of epileptic mice. This helps them identify behavioral fingerprints that often go unnoticed by humans.
It is a machine-learning technology that trains an unsupervised machine to identify repeated patterns of behavior. Once the behaviors are identified, the AI provides visualization tools and statistical tests that can help scientists understand and compare them to different experimental conditions.
This allows researchers to locate, track, and quantify mice behavior. Thus, they found that the technology could be better used to identify between epileptic and non-epileptic mice that outperforms trained human observers.
In addition, MoSeq only needed an hour of video recording and did not need a seizure to occur before it provides its analysis.
Also read: American Medical Association at CES 2023: How Digital Health and AI are Revolutionizing Patient Care
Better Epilepsy Diagnosis
The study shows the potential for better diagnosing epilepsy in humans. AI, specifically MoSeq can provide a faster and less labor intensive approach to diagnosis and treatment of epilepsy. MoSeq can also be used to monitor the progression of the disease and to identify any changes in the behavior that could indicate a seizure. This could give doctors access to more detailed information about a patient's condition that can be used to improve treatment.
How Motion Sequencing Works
Motion sequencing is a type of AI which uses computer vision and machine learning to identify patterns in video recordings of patients with epilepsy. It then uses this information to detect and classify seizure events in the video recording. The technology works by analyzing the movements of the patient and looking for changes in the movements that could indicate a seizure. It then uses this information to classify the seizure and provide an accurate diagnosis.
The study demonstrates the potential of AI to provide better diagnosis and care for patients with epilepsy. MoSeq is a promising technology that can provide a faster and less labor intensive approach to diagnosis and treatment. It can also be used to monitor the progression of the disease and to identify any changes in the behavior that could indicate a seizure. This could give doctors access to more detailed information about a patient's condition that can be used to improve treatment.
Related article: Diagnostic Robotics: Using AI to Detect Medical Problems, Tech Trained on 65M Health Records
