In recent years, more and more people are applying Convolutional Neural Networks to the study of sound signals. The main reason is the translational invariance of convolution in time and space. Thereby the diversity of the sound signal can be overcome. However, in terms of sound direction recognition, there are also problems such as a microphone matrix being too large, and feature selection. This paper proposes a sound direction recognition using a simulated human head with microphones at both ears. Theoretically, (...) the two microphones cannot distinguish the front and rear directions. However, we use the original data of the two channels as the input of the convolutional neural network, and the resolution effect can reach more than 0.9. For comparison, we also chose the delay feature (GCC) for sound direction recognition. Finally, we also conducted experiments that used probability distributions to identify more directions. (shrink)

Cranial electrical stimulation has been applied at various current levels in both adults and children with neurological conditions with seemingly promising but somewhat inconsistent results. Stimulation-induced spatial electric fields within a specific brain region are likely a significant contributing factor for the biological effects. Although several simulation models have been used to predict EF distributions in the brain, these models actually have not been validated by in vivo CES-induced EF measurements in the live human brain. This study directly measured the (...) CES-induced voltage changes with implanted stereotactic-electroencephalographic electrodes in twenty-one epilepsy participants and then compared these measured values with the simulated ones obtained from the personalized models. In addition, we further investigated the influence of stimulation frequency, intensity, electrode montage and age on EFs in parts of participants. We found both measured voltages and EFs obtained in vivo are highly correlated with the predicted ones in our cohort. In white matter and gray matter, the measured voltages linearly increased when the stimulation intensity increased from 5 to 500 μA but showed no significant changes with changing stimulation frequency from 0.5 to 200 Hz. Electrode montage, but not age, significantly affects the distribution of the EFs. Our in vivo measurements demonstrate that the individualized simulation model can reliably predict the CES-induced EFs in both adults and children. It also confirms that the CES-induced EFs highly depend on the electrode montages and individual anatomical features. (shrink)