Gender Differences Are Encoded Differently in the Structure and Function of the Human Brain Revealed by Multimodal MRI

Xi Zhang; Meng Liang; Wen Qin; Baikun Wan; Chunshui Yu; Dong Ming

doi:10.3389/fnhum.2020.00244

Gender Differences Are Encoded Differently in the Structure and Function of the Human Brain Revealed by Multimodal MRI

Front Hum Neurosci. 2020 Jul 21:14:244. doi: 10.3389/fnhum.2020.00244. eCollection 2020.

Authors

Xi Zhang^{1

2

3}, Meng Liang^{2

3}, Wen Qin^{3

4}, Baikun Wan¹, Chunshui Yu^{2

3

4}, Dong Ming^{1

5}

Affiliations

¹ Department of Biomedical Engineering, College of Precision Instruments and Optoelectronics Engineering, Tianjin University, Tianjin, China.
² School of Medical Imaging, Tianjin Medical University, Tianjin, China.
³ Tianjin Key Laboratory of Functional Imaging, Tianjin Medical University, Tianjin, China.
⁴ Department of Radiology, Tianjin Medical University General Hospital, Tianjin, China.
⁵ Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China.

Abstract

Despite widely reported gender differences in both brain structure and brain function, very few studies have examined the relationship between the structural differences and the functional differences between genders. Here, different imaging measures including both structural [i.e., gray matter volume (GMV)] and functional [i.e., regional homogeneity (ReHo) and functional connectivity (FC)] measures were employed to detect the gender differences in the human brain based on univariate and multivariate approaches with a sample of 290 healthy adults (155 females). The univariate analyses revealed that gender differences were detected in both structural (i.e., GMV) and functional (ReHo or FC) imaging measures, mainly manifested as greater values in females than in males in regions of the frontal, parietal, occipital lobes and cerebellum. Importantly, there was little overlap between the differences detected in GMV and those detected in ReHo and FC, and their differences between genders were not correlated with each other. The multivariate pattern analyses revealed that each of these measures had discriminative power to successfully distinguish between genders (classification accuracy: 94.3%, 90.73%, and 83.89% for GMV, ReHo, and FC, respectively) and their combination further improved the classification performance (96.6%). Our results suggest that gender differences are encoded in both brain structure and brain function, but in different manners. The finding of different and complementary information contained in structural and functional differences between genders highlights the complex relationship between brain structure and function, which may underlie the complex nature of gender differences in behavior.

Keywords: functional connectivity; gender difference; gray matter volume; multivariate pattern analysis; regional homogeneity; support vector machine.