小柯机器人

挪威科技大学Markus Frey、Matthias Nau等研究人员合作利用深度神经网络实现基于磁共振的眼球追踪。2021年11月8日，《自然—神经科学》杂志在线发表了这项成果。

为了使眼球追踪能够自由和广泛地用于磁共振成像研究，研究人员开发了DeepMReye，这是一个卷积神经网络（CNN），从眼球的磁共振信号中解码注视位置。它以亚成像的时间分辨率，在几乎没有训练数据的情况下和通用的扫描协议中，对保持不动的参与者进行无凸轮眼动跟踪。重要的是，它甚至在现有的数据集和闭眼时也能工作。解码的眼球运动解释了整个网络的大脑活动，也解释了与眼球运动功能无关的区域。

这项工作强调了眼动跟踪对解释fMRI结果的重要性，并提供了一个开源的软件解决方案，可广泛适用于研究和临床环境。

据了解，观察行为为了解人类认知和健康的许多核心方面提供了一个窗口，它是许多功能磁共振成像（fMRI）研究中感兴趣或混淆的一个重要变量。

附：英文原文

Title: Magnetic resonance-based eye tracking using deep neural networks

Author: Frey, Markus, Nau, Matthias, Doeller, Christian F.

Issue&Volume: 2021-11-08

Abstract: Viewing behavior provides a window into many central aspects of human cognition and health, and it is an important variable of interest or confound in many functional magnetic resonance imaging (fMRI) studies. To make eye tracking freely and widely available for MRI research, we developed DeepMReye, a convolutional neural network (CNN) that decodes gaze position from the magnetic resonance signal of the eyeballs. It performs cameraless eye tracking at subimaging temporal resolution in held-out participants with little training data and across a broad range of scanning protocols. Critically, it works even in existing datasets and when the eyes are closed. Decoded eye movements explain network-wide brain activity also in regions not associated with oculomotor function. This work emphasizes the importance of eye tracking for the interpretation of fMRI results and provides an open source software solution that is widely applicable in research and clinical settings. Viewing behavior is a key variable of interest but also a confound in fMRI studies. This paper presents a deep learning framework to decode gaze position from the magnetic resonance signal of the eyeballs, which enables eye tracking in fMRI data without a camera.

DOI: 10.1038/s41593-021-00947-w

Source: https://www.nature.com/articles/s41593-021-00947-w

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