Validation of fully automated total brain volumetric analysis by AccuBrain as compared to SIENAX for multiple sclerosis
Author(s): ,
A. Lau
Affiliations:
Department of Medicine and Therapeutics
,
J. Abrigo
Affiliations:
Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong
,
S. Lin
Affiliations:
Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong; BrainNow Medical Technology Ltd, Hong Kong, Hong Kong
,
Q. Chen
Affiliations:
Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong
,
C. Au
Affiliations:
Department of Medicine and Therapeutics
,
A. Ng
Affiliations:
Department of Medicine and Therapeutics
,
W. Chu
Affiliations:
Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong
,
S. Yu
Affiliations:
Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong
V. Mok
Affiliations:
Department of Medicine and Therapeutics
ECTRIMS Online Library. Lau A. Oct 12, 2018; 228871; P1029
Alexander Lau
Alexander Lau
Contributions
Abstract

Abstract: P1029

Type: Poster Sessions

Abstract Category: Clinical aspects of MS - Clinical assessment tools

Background: Brain volumetric analysis by MRI has become a crucial disease biomarker in multiple sclerosis (MS). AccuBrain (www.brainnow.cn) is a cloud-based system for fully-automated MRI brain region segmentation and measurements validated in neurodegenerative diseases, which can be further developed into a quick and reproducible tool for routine clinical use in MS.
Methods: Twenty volumetric 3D T1-weighted MRI brain scans from a single center were post-processed using Structural Image Evaluation using Normalisation of Atrophy-Cross-sectional (SIENAX), with additional modification of segmentation by manual lesion in-painting and removal of extracranial structures. Data on whole brain volumes (WBV) were obtained by AccuBrain and compared with SIENAX, which was used as the reference standard. Measures of reliability (Pearson's r for precision, bias correction factor Cbfor accuracy, % volume difference for discrepancy) were calculated. Statistical analysis was performed using MedCalc Statistical Software version 17.6 (MedCalc Software bvba, Ostend, Belgium; http://www.medcalc.org; 2017) with statistical significance set at p< 0.05.
Results: Of 20 subjects, the median age was 41.5 (interquartile range, 7.75) years. 14(70%) were female. The mean disease duration was 9.0 +/- 6.1 years. 18(90%) subjects were relapsing MS. The mean WBV by AccuBrain and SIENAX were 1020.2mL and 1028.3mL, respectively. There was excellent correlation between WBV measured by the two methods; Pearson's correlation was r=0.93 (95% CI 0.83-0.97, p< 0.001), bias correlation factor Cbwas 0.99, and % volume difference was 0.67% (range: -6.3% to +5.7%).
Discussion: AccuBrain is a precise (93%) and accurate (99%) tool to measure whole brain volume in a fully-automated fashion as compared with ground-truth measured by SIENAX, and does not require manual lesion in-painting for black holes and CSF spaces. Further validation with a larger sample size and external cohorts is warranted.
Disclosure: Acknowledgement: The study is supported by an educational grant by Novartis.
Disclosures: All authors have nothing to disclose.

By clicking “Accept Terms & all Cookies” or by continuing to browse, you agree to the storing of third-party cookies on your device to enhance your user experience and agree to the user terms and conditions of this learning management system (LMS).

Cookie Settings
Accept Terms & all Cookies