Osteoarthritis remains a major contributor to chronic pain, with currently available pharmacological treatments often proving limited and insufficient. A significant challenge in developing effective analgesics lies in accurately assessing pain behavior in preclinical rodent models. Recent advancements in machine learning algorithms and computational resources have enabled the automated detection and quantification of diverse behavioral parameters from video recordings of freely moving rodents. BlackBox imaging platform, capable of recording animal movements, body posture, and weight distribution across paws, has demonstrated efficacy in capturing and quantifying pain-related behaviors in various models. We validated the BlackBox platform for its capacity to automatically acquire and analyze reliable, quantitative, and objective data relating to spontaneous and unrestricted pain-like behaviors in rats. Several behavioral parameters were identified as altered in monoarthritic pain models, which were induced by complete Freund’s adjuvant (CFA) tibiotarsal or monosodium iodoacetate (MIA) knee intra-articular injections in rats. To evaluate the significance of these behavioral readouts in relation to pain phenotype, animals received a single intravenous injection of anti-NGF antibody (tanezumab), a clinically established analgesic proven effective in osteoarthritis pain management. We compared the performance of the BlackBox and CatWalk XT platforms in assessing static weight bearing and found that BlackBox also detected additional behavioral parameters responsive to tanezumab treatment. Both systems confirmed reduced weight bearing on the injured paw in CFA and MIA models—a deficit attenuated by tanezumab administration. Collectively, our findings strongly endorse the BlackBox system as an effective tool for quantifying pain behavior in rat osteoarthritis models.

Learning Objectives:

1. Explain how high-speed videography combined with machine learning enables automated detection and quantification of pain-like behaviors in preclinical rodent models.

2. Identify key behavioral parameters associated with osteoarthritis pain in CFA- and MIA-induced rat models as measured by the BlackBox imaging platform.

3. Evaluate the effectiveness of machine-learning–based behavioral analysis platforms (BlackBox vs. CatWalk XT) in detecting analgesic responses to anti-NGF (tanezumab) treatment.