Diagnostic Imaging Department, Fleni, Montañeses, 2325 (C1428AQK), Ciudad de Buenos Aires, Argentina.
Department of Physics, University of Buenos Aires (UBA), Buenos Aires, Argentina.
Eur Radiol. 2024 Mar;34(3):2024-2035. doi: 10.1007/s00330-023-10093-5. Epub 2023 Aug 31.
Evaluate the performance of a deep learning (DL)-based model for multiple sclerosis (MS) lesion segmentation and compare it to other DL and non-DL algorithms.
This ambispective, multicenter study assessed the performance of a DL-based model for MS lesion segmentation and compared it to alternative DL- and non-DL-based methods. Models were tested on internal (n = 20) and external (n = 18) datasets from Latin America, and on an external dataset from Europe (n = 49). We also examined robustness by rescanning six patients (n = 6) from our MS clinical cohort. Moreover, we studied inter-human annotator agreement and discussed our findings in light of these results. Performance and robustness were assessed using intraclass correlation coefficient (ICC), Dice coefficient (DC), and coefficient of variation (CV).
Inter-human ICC ranged from 0.89 to 0.95, while spatial agreement among annotators showed a median DC of 0.63. Using expert manual segmentations as ground truth, our DL model achieved a median DC of 0.73 on the internal, 0.66 on the external, and 0.70 on the challenge datasets. The performance of our DL model exceeded that of the alternative algorithms on all datasets. In the robustness experiment, our DL model also achieved higher DC (ranging from 0.82 to 0.90) and lower CV (ranging from 0.7 to 7.9%) when compared to the alternative methods.
Our DL-based model outperformed alternative methods for brain MS lesion segmentation. The model also proved to generalize well on unseen data and has a robust performance and low processing times both on real-world and challenge-based data.
Our DL-based model demonstrated superior performance in accurately segmenting brain MS lesions compared to alternative methods, indicating its potential for clinical application with improved accuracy, robustness, and efficiency.
• Automated lesion load quantification in MS patients is valuable; however, more accurate methods are still necessary. • A novel deep learning model outperformed alternative MS lesion segmentation methods on multisite datasets. • Deep learning models are particularly suitable for MS lesion segmentation in clinical scenarios.
评估基于深度学习（DL）的多发性硬化症（MS）病变分割模型的性能，并与其他 DL 和非 DL 算法进行比较。
这项前瞻性、多中心研究评估了基于 DL 的 MS 病变分割模型的性能，并与替代的 DL 和非 DL 方法进行了比较。模型在来自拉丁美洲的内部（n=20）和外部（n=18）数据集以及来自欧洲的外部数据集（n=49）上进行了测试。我们还通过重新扫描我们的 MS 临床队列中的六名患者（n=6）来检查稳健性。此外，我们还研究了人类注释者之间的一致性，并根据这些结果讨论了我们的发现。使用组内相关系数（ICC）、Dice 系数（DC）和变异系数（CV）评估性能和稳健性。
人类之间的 ICC 范围为 0.89 至 0.95，而注释者之间的空间一致性显示出中位数 DC 为 0.63。使用专家手动分割作为金标准，我们的 DL 模型在内部数据集上的中位数 DC 为 0.73，在外部数据集上为 0.66，在挑战数据集上为 0.70。在所有数据集上，我们的 DL 模型的性能均优于替代算法。在稳健性实验中，与替代方法相比，我们的 DL 模型在进行比较时也实现了更高的 DC（范围从 0.82 到 0.90）和更低的 CV（范围从 0.7 到 7.9%）。
我们的基于 DL 的模型在脑 MS 病变分割方面优于替代方法。该模型还在未见数据上表现出良好的泛化能力，并且在真实世界和基于挑战的数据上具有稳健的性能和低处理时间。
与替代方法相比，我们的基于 DL 的模型在准确分割脑 MS 病变方面表现出优异的性能，这表明其在提高准确性、稳健性和效率方面具有临床应用的潜力。
• 在 MS 患者中自动量化病变负荷具有重要价值；然而，仍需要更准确的方法。• 一种新的深度学习模型在多站点数据集上优于替代的 MS 病变分割方法。• 深度学习模型特别适合于临床情况下的 MS 病变分割。
