Validation of an automatic tool for the measurement of brain atrophy and white matter hyperintensity in clinical routine: QyScore®

Abstract

AbstractBackgroundBrain automatic segmentation methods have shown reliable results in measuring brain structure volumes and white matter lesions. Several studies showed that automated segmentation provides greater benefits than risks for patients affected by neurological diseases. However, few automated segmentation software are currently approved by regulatory agencies such as the US Food and Drug Administration (FDA). Here we present the methods and performance of QyScore®, an automated imaging analysis tool certified in Europe (CE marked) and US (FDA cleared) for the automatic volumetry of Grey and White Matter (GM and WM respectively) Hippocampus (HP), Amygdala (AM) and White Matter Hyperintensity (WMH).MethodA comprehensive database (N = 210) coming from several cohorts of patients with neurodegenerative diseases was established. The reliability measures considered for the validation of QyScore® imaging markers were: Dice Similarity Coefficient (DSC) and the Relative Volume Difference (RVD) for the HP, the AM, the GM and the WM volumes, DSC and the F1 metrics for the WMH. For each reliability index we identified thresholds and we hypothesized that DSC/F1 scores obtained using QyScore® markers were superior to the threshold, while RVD scores were inferior to the threshold. In addition, QyScore® performances were compared with the ones obtained from gold standard methods (manual segmentation) through regression analysis and Bland‐Altman plots.ResultComparison with reliability measures are described in Table 1. QyScore® compared with manual segmentation methods provides reliable performances for the segmentation of the grey and white matter volumes, hippocampal and amygdala volumes, as well as white matter hyperintensity volume as described in Figure 1.ConclusionQyScore® represents a reliable and user‐friendly tool, supporting the diagnosis made by clinicians for the early detection of neurological diseases such as MCI and AD dementia.