Gaubert Malo, Dell'Orco Andrea, Lange Catharina, Garnier-Crussard Antoine, Zimmermann Isabella, Dyrba Martin, Duering Marco, Ziegler Gabriel, Peters Oliver, Preis Lukas, Priller Josef, Spruth Eike Jakob, Schneider Anja, Fliessbach Klaus, Wiltfang Jens, Schott Björn H, Maier Franziska, Glanz Wenzel, Buerger Katharina, Janowitz Daniel, Perneczky Robert, Rauchmann Boris-Stephan, Teipel Stefan, Kilimann Ingo, Laske Christoph, Munk Matthias H, Spottke Annika, Roy Nina, Dobisch Laura, Ewers Michael, Dechent Peter, Haynes John Dylan, Scheffler Klaus, Düzel Emrah, Jessen Frank, Wirth Miranka
German Center for Neurodegenerative Diseases, Dresden, Germany.
Department of Neuroradiology, Rennes University Hospital (CHU), Rennes, France.
Front Psychiatry. 2023 Jan 12;13:1010273. doi: 10.3389/fpsyt.2022.1010273. eCollection 2022.
White matter hyperintensities (WMH), a biomarker of small vessel disease, are often found in Alzheimer's disease (AD) and their advanced detection and quantification can be beneficial for research and clinical applications. To investigate WMH in large-scale multicenter studies on cognitive impairment and AD, appropriate automated WMH segmentation algorithms are required. This study aimed to compare the performance of segmentation tools and provide information on their application in multicenter research.
We used a pseudo-randomly selected dataset ( = 50) from the DZNE-multicenter observational Longitudinal Cognitive Impairment and Dementia Study (DELCODE) that included 3D fluid-attenuated inversion recovery (FLAIR) images from participants across the cognitive continuum. Performances of top-rated algorithms for automated WMH segmentation [Brain Intensity Abnormality Classification Algorithm (BIANCA), lesion segmentation toolbox (LST), lesion growth algorithm (LGA), LST lesion prediction algorithm (LPA), pgs, and sysu_media] were compared to manual reference segmentation (RS).
Across tools, segmentation performance was moderate for global WMH volume and number of detected lesions. After retraining on a DELCODE subset, the deep learning algorithm sysu_media showed the highest performances with an average Dice's coefficient of 0.702 (±0.109 SD) for volume and a mean F1-score of 0.642 (±0.109 SD) for the number of lesions. The intra-class correlation was excellent for all algorithms (>0.9) but BIANCA (0.835). Performance improved with high WMH burden and varied across brain regions.
To conclude, the deep learning algorithm, when retrained, performed well in the multicenter context. Nevertheless, the performance was close to traditional methods. We provide methodological recommendations for future studies using automated WMH segmentation to quantify and assess WMH along the continuum of cognitive impairment and AD dementia.
白质高信号（WMH）是小血管疾病的生物标志物，在阿尔茨海默病（AD）中经常出现，其早期检测和定量分析对研究和临床应用有益。为了在大规模多中心认知障碍和AD研究中研究WMH，需要合适的自动WMH分割算法。本研究旨在比较分割工具的性能，并提供其在多中心研究中的应用信息。
我们使用了来自德国神经退行性疾病中心（DZNE）多中心纵向认知障碍和痴呆研究（DELCODE）的伪随机选择数据集（n = 50），其中包括来自认知连续体各阶段参与者的3D液体衰减反转恢复（FLAIR）图像。将自动WMH分割的顶级算法[脑强度异常分类算法（BIANCA）、病变分割工具箱（LST）、病变生长算法（LGA）、LST病变预测算法（LPA）、pgs和sysu_media]的性能与手动参考分割（RS）进行比较。
在所有工具中，全球WMH体积和检测到的病变数量的分割性能中等。在DELCODE子集中重新训练后，深度学习算法sysu_media表现出最高性能，体积的平均Dice系数为0.702（±0.109标准差），病变数量的平均F1分数为0.642（±0.109标准差）。除BIANCA（0.835）外，所有算法的类内相关性都非常好（>0.9）。随着WMH负担的增加，性能有所提高，并且在不同脑区有所差异。
总之，深度学习算法在重新训练后在多中心环境中表现良好。然而，其性能与传统方法相近。我们为未来使用自动WMH分割来量化和评估认知障碍和AD痴呆连续体中的WMH的研究提供了方法学建议。
