Tozlu Ceren, Ong Dylan, Piccirillo Christopher, Schwartz Hannah, Jaywant Abhishek, Nguyen Thanh, Jamison Keith, Gauthier Susan, Kuceyeski Amy
Res Sq. 2025 Apr 1:rs.3.rs-6214708. doi: 10.21203/rs.3.rs-6214708/v1.
Our prior work demonstrated that estimated structural and functional connectomes (eSC and eFC) generated using multiple sclerosis (MS) lesion masks and artificial intelligence (AI) models can predict disability as effectively as SC and FC derived from diffusion and functional MRI in MS. The goal of this study was to assess the ability of eSC and eFC in predicting baseline and 4-year follow-up cognition in MS patients. The Network Modification tool was performed to estimate SC from the clinical MRI-derived lesion masks. The eSC was then used as an input to Krakencoder, an encoder-decoder model, to estimate FC. The highest accuracy was obtained when predicting the follow-up Symbol Digit Modalities Test (SDMT) using regional eSC or eFC with a median Spearman's correlation of 0.58 for eSC and 0.56 for eFC, which is higher or similar to other studies that predicted cognition in healthy and diseased cohorts. Decreased eSC and eFC in the cerebellum and increased eFC in the default mode network were associated with lower follow-up SDMT scores. Our findings demonstrate that eSC and eFC derived from clinically acquired MRI and AI models can effectively predict cognition. The use of lesion-based estimates of connectome disruptions may potentially improve cognition-related individualized treatment planning.
我们之前的研究表明，使用多发性硬化症（MS）病变掩码和人工智能（AI）模型生成的估计结构和功能连接组（eSC和eFC）在预测残疾方面与从MS患者的扩散张量成像和功能磁共振成像中得出的SC和FC一样有效。本研究的目的是评估eSC和eFC在预测MS患者基线和4年随访认知方面的能力。使用网络修正工具从临床MRI得出的病变掩码中估计SC。然后将eSC用作编码器 - 解码器模型Krakencoder的输入来估计FC。当使用区域eSC或eFC预测随访符号数字模态测试（SDMT）时，获得了最高准确率，eSC的中位数斯皮尔曼相关性为0.58，eFC为0.56，这高于或类似于其他预测健康和患病队列认知的研究。小脑的eSC和eFC降低以及默认模式网络中的eFC增加与较低的随访SDMT分数相关。我们的研究结果表明，从临床获取的MRI和AI模型得出的eSC和eFC可以有效预测认知。基于病变的连接组破坏估计的使用可能会潜在地改善与认知相关的个性化治疗计划。
