Hippocampal ripples are transient neuronal features observed in high-frequency oscillatory bands of local field potentials, and they occur primarily during periods of behavioral immobility and slow-wave sleep. Ripples have been defined based on mathematically engineered features, such as magnitudes, durations, and cycles per event. However, the "ripples" could vary from laboratory to laboratory because their definition is subject to human bias, including the arbitrary choice of parameters and thresholds. In addition, local field potentials are often influenced by myoelectric noise arising from animal movement, making it difficult to distinguish ripples from high-frequency noises. To overcome these problems, we extracted ripple candidates under few constraints and labeled them as binary or stochastic "true" or "false" ripples using Gaussian mixed model clustering and a deep convolutional neural network in a weakly supervised fashion. Our automatic method separated ripples and myoelectric noise and was able to detect ripples even when the animals were moving. Moreover, we confirmed that a convolutional neural network was able to detect ripples defined by our method. Leave-one-animal-out cross-validation estimated the area under the precision-recall curve for ripple detection to be 0.72. Finally, our model establishes an appropriate threshold for the ripple magnitude in the case of the conventional detection of ripples.

中文翻译：

---

海马波纹的半监督定义

海马波纹是在局部场电势的高频振荡带中观察到的瞬时神经元特征，它们主要发生在行为不动和慢波睡眠期间。波纹是根据数学设计的特征定义的，例如幅度，持续时间和每个事件的周期。但是，“波纹”在实验室之间可能会有所不同，因为它们的定义会受到人为的偏见，包括参数和阈值的任意选择。另外，局部场电势通常受动物运动产生的肌电噪声的影响，因此很难区分波纹和高频噪声。为了克服这些问题，我们在很少的约束条件下提取了涟漪候选，并将其标记为二进制或随机的“ true”或“ false” 使用高斯混合模型聚类和深度卷积神经网络以弱监督的方式产生波纹。我们的自动方法可以分离出波纹和肌电噪声，即使在动物移动时也能够检测到波纹。此外，我们证实了卷积神经网络能够检测到由我们的方法定义的波纹。留一动物交叉验证估计用于脉动检测的精确召回曲线下的面积为0.72。最后，在常规检测波动的情况下，我们的模型为波动幅度确定了合适的阈值。此外，我们证实了卷积神经网络能够检测到由我们的方法定义的波纹。留一动物交叉验证估计用于脉动检测的精确召回曲线下的面积为0.72。最后，在常规检测波动的情况下，我们的模型为波动幅度确定了合适的阈值。此外，我们证实了卷积神经网络能够检测到由我们的方法定义的波纹。留一动物交叉验证估计用于脉动检测的精确召回曲线下的面积为0.72。最后，在常规检测波动的情况下，我们的模型为波动幅度确定了合适的阈值。