Sonix Health

Sonix Health is a Class II medical device designed to automate echocardiography quantification and reporting for licensed physicians, aiming to enhance efficiency and accuracy in cardiac image analysis. It employs a two-step artificial intelligence algorithm to process DICOM-compliant ultrasound images, perform automated measurements and strain analysis, and has demonstrated high accuracy and correlation with manual expert measurements in performance testing.

Ontact Health Co., Ltd has received a determination that its Sonix Health device is substantially equivalent to existing legally marketed devices. This allows the company to market Sonix Health, which is classified as a Class II medical image management and processing system. The device is intended for use by or on the order of a licensed physician for quantifying and reporting echocardiography, accepting DICOM-compliant ultrasound images from B (2D), M, Pulsed-wave Doppler, and Continuous-wave Doppler modes for adult populations.Sonix Health is a software-only solution designed to be installed directly on a customer's PC hardware. It utilizes a two-step artificial intelligence algorithm to automate tasks such as checking multiframe DICOMs, classifying echocardiography multiframe DICOMs, and performing automatic measurements, while also allowing for manual verification and adjustment of results, and providing analysis reports. Non-clinical and standalone testing confirmed its performance and compliance with relevant standards, including IEC 62304 for software life cycle processes and ISO 14971 for risk management. Performance tests on patient images demonstrated high accuracy for view recognition and strong correlation coefficients for automated measurements and strain analysis compared to manual expert measurements, thereby confirming its design specifications and substantial equivalence.

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. How does the two-step AI algorithm specifically handle complex or atypical echocardiography images that might challenge current manual interpretation?; What are the specific clinical scenarios or patient populations where Sonix Health's automated measurements and strain analysis show the most significant diagnostic or prognostic value?; Beyond view recognition and automated measurements, what other advanced analytical capabilities could be integrated into Sonix Health to further assist clinicians?; Given its software-only nature, what cybersecurity measures are in place to protect patient data and ensure the integrity of the AI algorithm on a customer's PC hardware?; What is the long-term plan for post-market surveillance and continuous improvement of the AI algorithm's accuracy, especially as new echocardiography techniques or pathologies emerge?