Clarius Median Nerve AI

The Clarius Median Nerve AI device is an assistive medical image processing system that uses a machine learning algorithm to perform semi-automatic, non-invasive measurement of the median nerve cross-sectional area from ultrasound data. Its purpose is to aid healthcare professionals by providing accurate measurements demonstrated to be non-inferior to manual expert measurements, thereby enhancing diagnostic workflows for conditions like carpal tunnel syndrome.

Clarius Mobile Health Corp.'s Clarius Median Nerve AI device has achieved substantial equivalence to a legally marketed predicate device, classifying it as a Class II medical image management and processing system. Its primary function is the segmentation and semi-automatic non-invasive measurement of the median nerve cross-sectional area using ultrasound data from Clarius Ultrasound Scanners. This assistive tool is intended for use by healthcare professionals trained in ultrasound on adult patients, with the user ultimately responsible for validating measurements based on clinical judgment. The device is subject to standard general controls, including annual registration, device listing, good manufacturing practices, and truthful labeling. The Clarius Median Nerve AI is integrated into the Clarius App software, leveraging a machine learning algorithm for automated detection, segmentation, and measurement. Its safety and effectiveness were rigorously established through comprehensive non-clinical and clinical performance testing, encompassing software verification and validation activities. A retrospective clinical verification study demonstrated that the device's automated measurements were non-inferior to those performed manually by human experts, exhibiting strong agreement. Furthermore, a clinical validation study confirmed the device's intended performance and its ability to meet user needs in a simulated environment. A Predetermined Change Control Plan (PCCP) has also been implemented to manage future modifications to the device's AI model, ensuring continued safety and effectiveness without requiring new premarket reviews if changes adhere to the plan.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific clinical conditions or patient demographics were included in the retrospective and validation studies, and how might these influence the device's applicability across diverse populations?; Beyond "strong agreement," what were the actual statistical metrics (e.g., Bland-Altman limits of agreement, intraclass correlation coefficient) demonstrating non-inferiority to manual measurements?; How does the Clarius Median Nerve AI device account for potential inter-operator variability in ultrasound probe placement or image acquisition, and what impact does this have on measurement consistency?; What are the criteria and review processes outlined in the Predetermined Change Control Plan (PCCP) to ensure ongoing safety and effectiveness for future AI model modifications?; Could the device's AI model be trained or adapted for other nerve measurements or pathologies, and what would be the regulatory pathway for such expansions?