PVAD IQ Software

PVAD IQ Software is a machine learning-based ultrasound image analysis tool designed to help clinicians position Percutaneous Ventricular Assist Devices (PVADs) by automatically detecting and measuring the distance between the aortic annulus and PVAD inlet in cardiac ultrasound images. It processes transthoracic echocardiography (TTE) clips from patients 18 years and older and provides clinical decision support through automated landmark detection and acceptability classification of ultrasound clips.

PVAD IQ uses deep convolutional neural networks implementing non-adaptive machine learning algorithms trained on clinical data to analyze ultrasound clips. It is substantially identical to the predicate device (LVivo Software) in its principle of operation, technology platform, algorithm type, anatomical site (heart), and scan type (vendor-neutral ultrasound). The key difference is that PVAD IQ measures distance between aortic annulus and PVAD inlet, whereas LVivo measures left ventricular ejection fraction and strain parameters.

Not stated in this summary. The document references FDA Guidance documents on "Content of Premarket Submissions for Device Software Functions" and "Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions" but does not cite specific ISO, IEC, or ASTM standards.

PVAD IQ is substantially equivalent to the LVivo predicate device because both share identical intended use (non-invasive ultrasound image processing), identical classification (Class II medical image management and processing system, product code QIH), identical principles of operation (machine learning-based ultrasound analysis), and identical algorithm type (deep convolutional neural networks). Although the specific measurements differ, these differences do not raise new safety or effectiveness questions, and PVAD IQ demonstrated equivalent performance through rigorous validation testing meeting pre-specified acceptance criteria.