小柯机器人

科学家开发出一个连接认知神经科学和人工智能的大规模7T fMRI数据集
2021-12-19 23:53
来源：小柯机器人

美国明尼苏达大学Kendrick Kay小组开发出一个连接认知神经科学和人工智能的大规模7T fMRI数据集。相关论文于2021年12月16日在线发表于国际学术期刊《自然—神经科学》。

研究人员提出了自然场景数据集（NSD），在参与者执行连续识别任务时，对大量自然场景的高分辨率功能磁共振成像反应进行了测量。为了优化数据质量，研究人员开发并应用了新的估计和去噪技术。对NSD数据的简单视觉检查揭示了沿腹侧视觉通路的清晰表征转换。研究人员利用NSD建立和训练了深度神经网络模型，比计算机视觉的最先进模型更准确地预测大脑活动，并进一步体现了该数据集的推理能力。NSD还包括大量的静止状态和扩散数据，使网络神经科学的观点能够制约和加强感知和记忆的模型。鉴于其前所未有的规模、质量和广度，NSD为认知神经科学和人工智能的研究开辟了新的途径。
 
据悉，在丰富的认知现象中对神经活动进行广泛的采样，对于有力地理解大脑功能至关重要。
 
附：英文原文

Title: A massive 7T fMRI dataset to bridge cognitive neuroscience and artificial intelligence

Author: Allen, Emily J., St-Yves, Ghislain, Wu, Yihan, Breedlove, Jesse L., Prince, Jacob S., Dowdle, Logan T., Nau, Matthias, Caron, Brad, Pestilli, Franco, Charest, Ian, Hutchinson, J. Benjamin, Naselaris, Thomas, Kay, Kendrick

Issue&Volume: 2021-12-16

Abstract: Extensive sampling of neural activity during rich cognitive phenomena is critical for robust understanding of brain function. Here we present the Natural Scenes Dataset (NSD), in which high-resolution functional magnetic resonance imaging responses to tens of thousands of richly annotated natural scenes were measured while participants performed a continuous recognition task. To optimize data quality, we developed and applied novel estimation and denoising techniques. Simple visual inspections of the NSD data reveal clear representational transformations along the ventral visual pathway. Further exemplifying the inferential power of the dataset, we used NSD to build and train deep neural network models that predict brain activity more accurately than state-of-the-art models from computer vision. NSD also includes substantial resting-state and diffusion data, enabling network neuroscience perspectives to constrain and enhance models of perception and memory. Given its unprecedented scale, quality and breadth, NSD opens new avenues of inquiry in cognitive neuroscience and artificial intelligence. The authors measured high-resolution fMRI activity from eight individuals who saw and memorized thousands of annotated natural images over 1 year. This massive dataset enables new paths of inquiry in cognitive neuroscience and artificial intelligence.

DOI: 10.1038/s41593-021-00962-x

Source: https://www.nature.com/articles/s41593-021-00962-x

Nature Neuroscience：《自然—神经科学》，创刊于1998年。隶属于施普林格·自然出版集团，最新IF：28.771
官方网址：https://www.nature.com/neuro/
投稿链接：https://mts-nn.nature.com/cgi-bin/main.plex


本期文章：《自然—神经科学》：Online/在线发表

分享到:

0