NIH grant: Optimizing fluorescent nanoparticle-based motion capture to understand the neural control of movement in health and disease (1R01AR087123-01A1)

retained text

NIH grant 1R01AR087123-01A1: Optimizing fluorescent nanoparticle-based motion capture to understand the neural control of movement in health and disease
Institute: {'code': 'AR', 'abbreviation': 'NIAMS', 'name': 'National Institute of Arthritis and Musculoskeletal and Skin Diseases', 'admin_org_id': '12500006', 'admin_funding_url': ''}. PI: Jeffrey Evan Markowitz. Award: $453560.
Start: 2026-08-21T00:00:00.
Project summary/abstract
 Quantifying movement in laboratory animals is central to the study of neuroscience. Recent advances
in machine learning have catalyzed a revolution in how movement is measured in the lab. Unlike
humans, laboratory animals cannot easily don motion capture suits bearing markers, spurring the
development of “markerless” motion tracking methods. Machine learning models have been trained to
identify key body parts without physical fiducials. However, the precision of markerless tracking in mice remains
two orders of magnitude worse (5-10 mm) than the gold-standard marker-based motion capture systems
used in humans (0.1 mm). Consequently, markerless keypoint trackers are limited in their ability to capture
fine-grained kinematics, which are essential for identifying ethologically relevant behaviors like self-grooming,
skilled reaching, and locomotion. Finally, markerless keypoint systems are limited to surface tracking and
cannot directly capture the primary output of the brain's motor control circuits: the musculoskeletal system.
This constitutes a key technological gap that has limited the neuroscience community's ability to investigate the
neural control of movement.
 To bridge this gap, the Markowitz lab has developed a new technology that leverages biocompatible
near-infrared fluorescent nanoparticles to form discrete long-lived deposits beneath the skin, which can
be non-invasively imaged as fiducial markers in freely moving animals using standard cameras. This grant
application involves the mechanistic characterization and further advancement of this technology, called
Quantum Dot-based Pose estimation in vivo (QD-Pi), to make it accessible to the neuroscience community.
Aim 1 focuses on increasing the longevity of in vivo fluorescent markers by attaching the fluorescent
nanoparticles to larger structures, such as biocompatible microbeads, thereby reducing nanoparticle
dispersal and clearance. Extending imaging timelines to multiple months will enable experiments that aim to
characterize the development of progressive movement disorders. Aim 2 focuses on targeting the fluorescent
nanoparticles to specific body parts (muscles, tendons, joints, bladder, stomach, whiskers) to support in
vivo motion tracking of specific physiological systems. This is feasible because the fluorescent nanoparticles
can be conjugated to highly specific targeting molecules (e.g. antibodies and peptides). Finally, Aim 3
focuses on the development of machine learning algorithms to automatically label video of fluorescent tags
with identified body parts, opening up the ability to amass ground-truth motion tracking datasets of
unprecedented size for benchmarking and training new keypoint trackers.

document fetched 2026-08-30 02:46