Analysis of complex and subtle behavior enabled by self-supervised deep learning
Studying freely moving animals is essential for understanding natural behaviors such as locomotion, foraging, or social interactions. Recent advancements in cameras, motion capture, and pose estimation have enabled high-throughput analysis of animal movement. While deep learning (DL) methods are widely used for human movement analysis and sequential data, their application to animal behavior remains is still emerging, primarily for pose estimation.
Motion capture and pose estimation provide high-quality data on individual movement but often contain missing values that hinder downstream analysis. To address this, I developed Deep Imputation for Skeleton Data (DISK), a deep learning algorithm that reconstructs missing tracking data by leveraging spatial and temporal dependencies between keypoints. We demonstrated its effectiveness across species and behaviors and developed it into a user-friendly tool with an uncertainty score to assess imputation quality (github.com/bozeklab/DISK.git).
My research focuses on developing flexible DL methods to link animal movement with neural activity, utilizing unsupervised and transfer learning approaches to minimize manual labeling. I aim to investigate how biological perturbations-genetic, pharmacological, and social-modulate behavior, ultimately creating a framework to enhance our understanding of behavioral dynamics. A precise, data-driven description of natural, unconstrained behavior could also inform the diagnosis and treatment of musculoskeletal and neurological disorders.
Short Bio: France ROSE is a post-doctoral researcher at the University Hospital of Cologne. With a strong background in computational biology, data processing, and machine learning, she is interested in the complexity of animal behavior and its link to neural activity, leveraging state-of-the-art deep learning methods.
France holds a Ph.D. in Computational Biology from the Institut de Biologie de l'École Normale Supérieure (IBENS, Paris). Before joining the University of Cologne, she worked as a data scientist at the AI-driven mental health start-up, MyndBlue, in Paris. Since August 2023, France has been self-funded through a "KI-starter" grant from the German state of NRW, and she has recently been awarded the prestigious Emmy Noether starting grant from the DFG. Outside of her academic work, she is passionate about gender equality in STEM and founded Coding Sisters, an outreach initiative supporting girls and gender minorities in coding. She is also a lyrical singer, and enjoys outdoor activities such as hiking, climbing and snorkeling.
more about Frances Rose is available at: http://www.normalesup.org/~frose/