{"id":"07566","slug":"autoomr-automated-optomotor--07566","source":{"id":"07566","dataset":"techtransfer","title":"AutoOMR: Automated Optomotor Reflex Behavior Detection","description_":"<p>University of Pittsburgh researchers have developed a novel, low-cost device for automated detection, classification and quantification of Optomotor Reflex (OMR) behavior in animals. Using digital cameras and machine learning AutoOMR can automate behavioral assay pipelines through standardization of data collection, interpretation of movements, event calling and quantitative reporting in OMR behavior. This novel approach could revolutionize the use of OMR behavior in research and make the technique widely accessible, scalable and reproducible. For preclinical programs — including retinal gene therapy, neuroprotection, and drug efficacy studies - AutoOMR offers a standardized, reproducible functional endpoint suitable for data intended for regulatory review, and consistent enough to support multi-site studies.</p><p><h2>Description</h2>OMR behavior is commonly used in research to assess visual function and nervous system in different animals. The process involves detecting, classifying and quantifying the animal’s response to various visual stimulations. Currently, OMR testing requires costly specialist systems often with subjective human observation with manual grading, limiting the accessibility, scalability, and reliability of the technique. AutoOMR overcomes these limitations, automating and standardizing OMR analysis in research. Because OMR depends on the full visual-motor pathway rather than the retina alone, changes in optomotor response can also serve as an early functional biomarker in neurodegenerative disease models, broadening AutoOMR's relevance beyond ophthalmology into neuroscience drug development.</p><p><h2>Applications</h2>- Visual testing\r<br>- Preclinical disease models\r<br>- Drug efficacy and safety testing\r<br>- Gene therapy endpoint assessment \r<br>- Neurodegenerative disease modeling (Parkinson's, Alzheimer's, TBI) via oculomotor biomarkers\r<br>- Behavioral and therapeutic response analysis</p><p><h2>Advantages</h2>AutoOMR uses overhead video to capture physical responses (e.g., ear or tail orientation) to a visual stimulus. A novel algorithm extracts meaningful physical responses to classify and quantify OMR behavior removing the need for subjective human intervention. AutoOMR is designed to only recognize true optomotor responses, excluding unrelated body movement, resulting in more sensitive OMR behavioral analysis. This novel automated approach is low-cost, using a  commercially available webcam to capture ear and tail positions, and can be easily integrated into preclinical models to provide standardized, quantified, OMR behavior analysis without the need for specialist equipment. Additionally, this approach removes the subjective human observation element of OMR testing ensuring results are reproducible and the technique is more widely available to researchers.</p><p><h2>Invention Readiness</h2>AutoOMR was developed consisting of a stimulus generator, a modular animal stage and a computer-vision unit. Wildtype and visually impaired mice were exposed to rotating black and white gratings at various directions, spatial frequency, speed and contrast. The webcam was used to record animal response to various visual stimuli. Using 14,286 sequences split into training, validation, and testing sets it was confirmed this approach could accurately quantify OMR behavior.</p><p><h2>IP Status</h2>Patent Pending</p><p><h2>Related Publication(s)</h2><p>Shaohua Pi, Justin Chen, Kevin Chen, Lingyun Wang, Bingjie Wang, Jose Alain Sahel; AutoOMR: A Deep Learning-Based Automated Platform for Quantifying Optomotor Reflex as a Visual Acuity Endpoint. Invest. Ophthalmol. Vis. Sci. 2026;67(7):2445. </p></p>","tags":["Life Science"],"file_number":"07566","collections":[],"meta_description":"Automated optomotor reflex analysis using webcam and deep learning for standardized, scalable, cost-efficient behavioral endpoint.","image_url":"","apriori_judge_output":"{\"scores\":{\"novelty\":4.0,\"potential_impact\":4.0,\"readiness\":4.0,\"scalability\":4.0,\"timeliness\":3.0},\"weighted_score\":3.9,\"risks\":[\"Potential overlap with existing OMR methods\",\"Regulatory acceptance uncertainties for animal behavior endpoints\",\"Need for broader validation across species and settings\",\"IP status and freedom-to-operate considerations pending patent-pending\",\"Dependence on video quality and standardized protocols\"],\"one_sentence_take\":\"High novelty and readiness with strong impact potential, but marginal timeliness given patent-pending status and need for broader regulatory validation.\"}","lead_inventor_name":"Shaohua Pi","lead_inventor_dept":"Med-Ophthalmology","technology_type":"Life Science Research Tool","technology_subtype":"Other Life Science Research Tool","therapeutic_areas":["Neuroscience","Ophthalmology"],"therapeutic_indications":[],"custom_tags":[],"all_tech_innovators":["Shaohua Pi","Lingyun Wang"],"date_submitted":"2026-04-21"},"highlight":{},"matched_queries":null,"score":0.0}