Institute of Medical Science and Technology, Shahid Beheshti University, Tehran, Iran.
Clinical and Counselling Psychology Program, Adler University, Vancouver, British Columbia, Canada.
Hum Brain Mapp. 2019 Dec 1;40(17):5142-5154. doi: 10.1002/hbm.24746. Epub 2019 Aug 4.
Over the past decades, neuroimaging has become widely used to investigate structural and functional brain abnormality in neuropsychiatric disorders. The results of individual neuroimaging studies, however, are frequently inconsistent due to small and heterogeneous samples, analytical flexibility, and publication bias toward positive findings. To consolidate the emergent findings toward clinically useful insight, meta-analyses have been developed to integrate the results of studies and identify areas that are consistently involved in pathophysiology of particular neuropsychiatric disorders. However, it should be considered that the results of meta-analyses could also be divergent due to heterogeneity in search strategy, selection criteria, imaging modalities, behavioral tasks, number of experiments, data organization methods, and statistical analysis with different multiple comparison thresholds. Following an introduction to the problem and the concepts of quantitative summaries of neuroimaging findings, we propose practical recommendations for clinicians and researchers for conducting transparent and methodologically sound neuroimaging meta-analyses. This should help to consolidate the search for convergent regional brain abnormality in neuropsychiatric disorders.
在过去的几十年中，神经影像学已广泛用于研究神经精神疾病的结构和功能脑异常。然而，由于样本量小且异质、分析的灵活性以及对阳性发现的发表偏倚，个别神经影像学研究的结果常常不一致。为了将新兴的发现整合为具有临床意义的见解，已经开发了荟萃分析来整合研究结果，并确定与特定神经精神疾病的病理生理学一致的区域。然而，应该考虑到荟萃分析的结果也可能因搜索策略、选择标准、成像方式、行为任务、实验数量、数据组织方法以及具有不同多重比较阈值的统计分析的异质性而存在分歧。在介绍问题和神经影像学发现的定量总结概念之后，我们为临床医生和研究人员提出了进行透明且方法合理的神经影像学荟萃分析的实用建议。这有助于整合神经精神疾病中趋同的区域性大脑异常的研究。
