LiverMultiScan v5 (LMSv5)

LiverMultiScan v5 is a post-processing software application that analyzes magnetic resonance images to non-invasively quantify liver tissue characteristics, specifically Proton Density Fat Fraction (PDFF), T2*, Liver Iron Concentration (LIC), and iron-corrected T1 (cT1). It employs automated segmentation, region of interest analysis, and parametric mapping with artefact detection to produce a quantitative report aiding clinicians in liver assessment.

LiverMultiScan v5 (LMSv5) is a standalone software device designed for the non-invasive evaluation of liver tissue characteristics by post-processing Magnetic Resonance (MR) images. It utilizes DICOM 3.0 compliant MR image datasets from compatible systems to generate, display, and review 2D medical images and pixel maps representing MR relaxation times. The software includes tools such as automated liver segmentation leveraging machine-learning, region of interest (ROI) placement, and automatic artefact detection to assist a trained operator in quantifying metrics from parametric maps. The primary purpose is to provide quantitative assessment, and a summary report is generated for interpretation by a clinician as an aid to diagnosis, without replacing established clinical procedures or making direct diagnostic decisions. LMSv5 quantifies several key liver metrics: triglyceride fat fraction (Proton Density Fat Fraction, PDFF), T2*, Liver Iron Concentration (LIC), and iron-corrected T1 (cT1). PDFF is quantified using the LMS IDEAL method, while LIC is derived from measured T2* values transformed by a calibration curve. The cT1 map is generated by correcting T1 measurements for the effects of elevated iron and standardizing for fat signals. The software visualizes these metrics through parametric maps and allows for quantification via full liver segmentation or manual ROI placement, reporting median and interquartile range (IQR) metrics. It is intended to provide additional information to clinicians as part of a comprehensive diagnostic process for patients presenting with symptoms or risk factors of liver disease.

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 is the accuracy and reproducibility of the Liver Iron Concentration (LIC) derived from T2* values compared to direct R2-based methods or liver biopsy, especially at varying levels of iron overload?; How does the machine-learning based automatic liver segmentation perform in patients with atypical liver morphology or extensive focal lesions, and what is the typical operator intervention rate?; Can the system differentiate between various causes of elevated cT1 values beyond iron and fat, and what is its sensitivity to subtle fibrotic changes?; What is the impact of MR scanner field strength and vendor variations on the quantitative metrics (PDFF, T2*, LIC, cT1), and what specific calibration or normalization steps are implemented in LMSv5 to ensure consistency?; How does the automatic artefact detection feature influence the selection of image slices for quantification, and what is its performance in identifying clinically significant artefacts that could alter diagnostic interpretation?