There is a critical need for more accurate neurological rating scales, as current scales are outdated, lack validation, and suffer from high inter-rater variability. Improved rating methods are essential for better clinical decision-making, more efficient therapy development, and the growing demand for objective digital assessments in telehealth and decentralized trials. Dannce.ai utilizes deep learning technology called DANNCE, which analyzes patient videos to track 3D poses and behaviors, providing valuable insights into neuropsychiatric health. This innovative approach, supported by patent protection and a team of experts in quantitative behavioral analysis, has demonstrated its effectiveness in Parkinson's patients and in preclinical drug phenotyping and disease diagnosis, with over 10,000 hours invested in its development. It offers a noninvasive means of capturing rich behavioral data for clinicians and drug developers, enabling them to uncover previously hidden information.
What is the problem?
Drug developers, clinicians, and patients need more accurate rating scales for neurological testing. It is well established that disability scales for these conditions, many developed over 40 years ago and never validated, are limited by their sensitivity and usability. Most neurologic patients' symptoms are rated on a "0-4" scale that has high inter-rater variability due to the subjective nature of scoring. High variance in scoring not only leads to impaired clinical decision making but also makes it challenging to determine the efficacy of therapies undergoing clinical development, forcing companies to run much larger trials in order to separate signal from noise. A better understanding of patient response to treatment would help pharma/biotech shelve "bound to fail" programs earlier in development, avoiding costly late-stage failures. In conjunction, the rise of telehealth and decentralized trials is increasing the need for automated and objective digital methods to measure neurological state.
What is their solution?
dannce.ai will help clinicians see the unseeable and drug developers generate real world evidence, by using movement data and computer vision. Their deep learning technology, DANNCE, takes patient video as input and uses their proprietary 3D pose tracking and behavioral analysis algorithms to quantify behavior. Behavior is a result of the complex coordination of brain processes and is affected by disease or impairment. As a result, the behavioral range of a patient or animal is a rich noninvasive source of information about its neuropsychiatric health. They have a unique and patent protected technical approach that has high scientific rationale, backed by a team of world experts at quantitative behavioral analysis. Their approach took over 10,000 hours to develop, has human proof of concept in Parkinson's patients, and an extensive preclinical track record of phenotyping drug effects and diagnosing disease.
