Brian Q. Geuther

Optimizing Intermediate Representations: A Framework for Low-Cost, High-Accuracy Behavior Quantification

This paper benchmarks intermediate representations for mouse behavior classification, finding that performance is robust to keypoint variation and temporal features improve accuracy, with segmentation matching pose estimation.

Visual detection of seizures in mice using supervised machine learning

This study uses supervised machine learning to automatically quantify seizure severity from video data in mice, enabling high-throughput and standardized seizure scoring.

Highly Accurate and Precise Determination of Mouse Mass Using Computer Vision

This paper demonstrates a computer vision method to estimate mouse body mass from video data with 4.8% error, enabling noninvasive and continuous monitoring.

Action detection using a neural network elucidates the genetics of mouse grooming behavior

The paper develops a neural network for automated mouse grooming behavior detection and uses it to study the genetic basis of grooming and its links to psychiatric traits.