LVivo Seamless

LVivo Seamless v2.0 is an offline medical image processing software that automates the analysis of adult echocardiograms by selecting relevant DICOM clips and utilizing its LVivo EF and LVivo Strain modules to calculate Ejection Fraction and Global Longitudinal Strain. This system aims to provide quantitative cardiac function parameters to healthcare professionals via PACS integration, aiding in the assessment of patients with suspected heart disease.

LVivo Seamless v2.0 is a standalone software application designed for the non-invasive processing of cardiac ultrasound (echocardiography) images for adult patients (age >18) with suspected heart disease. It functions as an extension of the LVivo Platform, operating offline on a server within a healthcare environment. The system receives echocardiographic examinations in DICOM format from ultrasound devices and employs an automated methodology to select appropriate image clips crucial for evaluating Ejection Fraction (EF) and Global Longitudinal Strain (GLS). Following the automated selection of cardiac views, LVivo Seamless utilizes its integrated LVivo EF and LVivo Strain modules to perform an automatic analysis of cardiac structure and function. The underlying science involves sophisticated image processing algorithms to identify cardiac structures, measure volumes (with features like Simpson's method of discs for volume calculation), and quantify parameters such as EF and strain. Performance data from a study of 100 patient examinations indicated that the system correctly identified 4CH, 2CH, and 3CH views in 95% of cases. Its GLS measurements demonstrated a strong correlation (0.88) and a mean difference of -1.4% (±3.93%) when compared to a reference Automated Functional Imaging (AFI) method, achieving a sensitivity of 0.90 and specificity of 0.83 for detecting abnormal GLS. The quantitatively analyzed results are subsequently sent to a Picture Archiving and Communication System (PACS) for review and interpretation by healthcare professionals.

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 LVivo Seamless algorithm for automated clip selection handle variations in image quality or atypical cardiac anatomies, and what is the impact on the 95% view recognition accuracy in such scenarios?; What are the specific algorithmic advancements in the LVivo Strain module for GLS calculation per view, as compared to the previous derivation of GLS from the LVivo EF module in the predicate device?; Considering the Bland-Altman analysis for GLS showing a mean difference of -1.4% (±3.93%) against AFI, what are the clinical implications for patients whose GLS values fall near the diagnostic threshold of -18%?; The validation study included patients with existing AFI GLS results; how is the performance of LVivo Seamless anticipated to vary in a more general patient population where such comparator data might be absent or of differing quality?; While the system provides automated EF and GLS evaluations, what specific manual editing capabilities are available for the output results, and how does the system manage or log these user modifications before finalization in PACS?