基于深度学习的大脑性别差异分析

Analysis of gender differences in the brain based on deep learning

深度学习被广泛应用于大脑的相关研究中。通过构建深度学习模型对弥散张量成像数据的各向异性分数进行了性别分类,并通过深度学习特征可视化方法提取了不同性别的重要特征,最后对可视化结果进行了基于体素的分析。结果显示,提出的模型能够准确预测性别,并且达到了96.2%的分类准确率。在可视化的结果中,发现男女大脑之间存在明显差异,其中存在差异的脑区主要表现在胼胝体、顶叶下叶和基底神经节等,这些脑区揭示了男女之间的大脑差异可能与运动能力、数学运算、身体形象感知和情绪控制等方面的能力相关。

Deep learning is widely used in brain related research.A deep learning model was constructed to classify the fraction anisotropy of the diffusion tensor imaging data.And the important features of different genders were extracted through the deep learning feature visualization method.Finally the visualization results were analyzed based on voxels.The results show that the proposed model can accurately predict gender and achieve a classification accuracy of 96.2%.In the visualized results,it is found that there are obvious differences between the brains of men and women.The brain regions with differences are mainly manifested in the corpus callosum,inferior parietal lobule and basal ganglia.These brain regions reveal that the brain differences between men and women may be related to exercise ability,mathematical operations,body image perception,and emotional changes.