Rai Shefali, Godfrey Kate J, Graff Kirk, Tansey Ryann, Merrikh Daria, Yin Shelly, Feigelis Matthew, Demeter Damion V, Vanderwal Tamara, Greene Deanna J, Bray Signe
Child and Adolescent Imaging Research Program, University of Calgary, Calgary, AB, Canada.
Alberta Children's Hospital Research Institute, University of Calgary, Calgary, AB, Canada.
Imaging Neurosci (Camb). 2025 Aug 19;3. doi: 10.1162/IMAG.a.117. eCollection 2025.
Reliable functional connectivity (FC) measurements are important for robust neuroimaging findings, yet pediatric functional magnetic resonance imaging (fMRI) faces unique challenges due to head motion and bias toward shorter scans. Passive viewing conditions during fMRI offer advantages for scanning pediatric populations, but FC reliability under these conditions remains underexplored. Here, we used precision fMRI data collected across three passive viewing conditions to directly compare FC reliability profiles between 25 pre-adolescent children and 25 adults, with each participant providing over 2.8 hours of data over 4 sessions. We found that FC test-retest correlations increased asymptotically with scan length, with children requiring nearly twice the post-censored scan time (24.6 minutes) compared with adults (14.4 minutes) to achieve comparable reliability, and that this effect was only partly attributable to head motion. Reliability differences between lower-motion adults and higher-motion children were spatially non-uniform and largest in ventral anterior temporal and frontal regions. While averaging features within functional networks improved intraclass correlation coefficient (ICC) reliability, values for higher-motion children remained in the poor-to-fair ICC range even with 24 minutes of data. Of note, we observed substantial increases in edge-wise ICC between 24 and 54 minutes of data. Viewing conditions with greater engagement reduced head motion in children but had lower FC reliability than less engaging "low-demand" videos, suggesting complex state- or condition-related trade-offs. These findings have important implications for developmental neuroimaging study design, particularly for higher motion pediatric populations.
可靠的功能连接（FC）测量对于获得稳健的神经影像学研究结果很重要，然而，由于头部运动以及倾向于进行更短扫描时间的扫描，儿科功能磁共振成像（fMRI）面临着独特的挑战。fMRI期间的被动观看条件为扫描儿科人群提供了优势，但在这些条件下FC的可靠性仍未得到充分探索。在这里，我们使用了在三种被动观看条件下收集的精确fMRI数据，直接比较了25名青春期前儿童和25名成年人之间的FC可靠性概况，每位参与者在4次扫描中提供了超过2.8小时的数据。我们发现，FC重测相关性随着扫描长度渐近增加，与成年人（14.4分钟）相比，儿童需要近两倍的经删失后扫描时间（24.6分钟）才能达到可比的可靠性，并且这种影响仅部分归因于头部运动。低运动的成年人与高运动的儿童之间的可靠性差异在空间上是不均匀的，并且在腹侧颞叶前部和额叶区域最大。虽然对功能网络内的特征进行平均提高了组内相关系数（ICC）的可靠性，但即使有24分钟的数据，高运动儿童的ICC值仍处于较差到中等的范围内。值得注意的是，我们观察到在24至54分钟的数据之间，逐边ICC有显著增加。参与度更高的观看条件减少了儿童的头部运动，但与参与度较低的“低需求”视频相比，FC可靠性较低，这表明存在复杂的与状态或条件相关的权衡。这些发现对发育神经影像学研究设计具有重要意义，特别是对于头部运动较多的儿科人群。
