6440 MyLabX90

The MyLabX90 system is an advanced ultrasound imaging platform designed to enhance diagnostic capabilities across various clinical applications by integrating established ultrasound modalities with novel artificial intelligence-powered tools for cardiac analysis and breast lesion characterization. Its methodology combines versatile hardware, diverse imaging modes, and sophisticated software options like Virtual Navigator for image fusion and quantitative analysis features to provide clinicians with comprehensive and precise information.

The MyLabX90 is a multifunctional, mobile ultrasound scanner designed for collecting, displaying, and analyzing ultrasound images across a wide range of clinical applications, including cardiac, vascular, general imaging, and women's health. A key feature is the Virtual Navigator option, which enhances radiological ultrasound examinations by integrating image information from secondary modalities like CT, MR, US, PET, XA, and NM, thereby providing additional anatomical context. The system supports various imaging modes such as B-Mode, M-Mode, Doppler, 3D/4D, and qualitative elastosonography (ElaXto), and is compatible with multiple probe types including phased array, convex, linear, and volumetric probes, facilitating its use in diverse clinical settings from examination rooms to operating rooms. This upgraded system introduces several advanced capabilities, notably leveraging artificial intelligence. The Cardio Package includes Auto E.F., an AI tool powered by Pie Medical Imaging's Caas Qardia software, for automatic left ventricular volume and ejection fraction calculation. The eDetect for Breast Lesions feature employs AI for contouring suspicious lesions and suggesting BI-RADS parameters. XStrain technology, also AI-powered, quantifies myocardial mechanics like strain and strain rate. Other enhancements include Quality Attenuation Imaging (QAI) for real-time quantitative tissue analysis, a Prostate Biopsy Stepper for guided procedures, HyperDoppler for advanced intracardiac flow visualization, new transducers, and an improved HDR display. Validation testing for AI components, such as breast lesion contouring (achieving an Intersection over Union > 0.85) and endocardial border segmentation (Dice coefficient 0.95), demonstrated performance aligned with predefined acceptance criteria using independent test datasets.

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 performance of the AI-driven Auto E.F. compare to manual ejection fraction calculations by experienced cardiologists across diverse patient populations with varying cardiac pathologies?; What is the learning curve and inter-operator variability associated with using the eDetect for Breast Lesions feature, and how does its BI-RADS parameter suggestion impact diagnostic accuracy and workflow efficiency?; Can the XStrain feature reliably differentiate between various cardiomyopathies or detect subclinical cardiac dysfunction, and how does its AI-powered border detection handle images of suboptimal quality?; What specific tissue characteristics or pathologies are best differentiated by the QAI (Quality Attenuation Imaging) feature, and how does its real-time quantitative analysis correlate with established quantitative ultrasound techniques or histopathology?; For the Virtual Navigator, what is the typical registration accuracy achieved when fusing ultrasound with different second modalities (CT, MR, PET), and how does this impact the precision of image-guided interventions like percutaneous procedures or prostate biopsies?