Results suggest new technology enables improved, high-volume characterization and quantification of in vivo behavior. Researchers at Boston Children’s Hospital and Harvard Medical School’s Department of Neurobiology have developed a new approach to measure dynamic changes in behavior in an unbiased, observer-independent manner, according to a new study published in and featured on the cover of PAIN, the journal of the International Association for the Study of Pain (IASP). The investigators noted that this technology platform can characterize a broad range of physiological and pathological phenomena and can be applied to objectively and automatically characterize behaviors including tactile hyperalgesia, sedation, and other neurological readouts.

“The application of machine learning to accurately capture freely moving voluntary rodent behavior and the analysis of this by machine learning provides the opportunity to detect subtle but important reflections of how it responds and how its nervous system is operating, as well as the presence of any disturbance in the nervous system, all in an unbiased automatic fashion.” Principal Investigator, Clifford Woolf, Mb, BCh, PhD, Professor of Neurobiology at Harvard Medical School and Director of the F.M. Kirby Neurobiology Center at Boston Children’s Hospital.

This peer-reviewed study demonstrated the efficacy of a novel automated technology platform in recording freely behaving mice over time for continuous data acquisition. The patented device technology has two parallel video data streams: a near-infrared frustrated total internal reflection (FTIR) for detecting the degree, force, and timing of surface contact, and near-infrared trans-illumination for ongoing video graphing of whole-body pose. In addition, researchers utilized a proprietary machine learning software for automatic extraction and quantification of behavior characters, allowing for objective, sensitive, and high-throughput measurement of the behavioral state of rodents.

The new approach seeks to solve for the current lack of sensitive and robust behavioral assessments of pain in preclinical models, a major limitation for both pain research and the development of novel analgesics.

“Historically behavioral assays have relied on reflex-withdrawal, and are commonly limited by only capturing snapshots of pain dynamics and requiring extensive animal-human observer interaction,” said study initiator David Roberson, a former postdoctoral fellow under Dr. Woolf who is now bringing the technology to market as co-founder and CEO of Blackbox Bio. “We’re incredibly excited to work with researchers around the world to make the technology an indispensable tool, not only for improving the efficacy of analgesics and other neurology medicines, but also in the safety evaluation of virtually any new medicine.”