Background: Automatic diagnosis of depression based on speech can complement mental health treatment methods in the future. Previous studies have reported that acoustic properties can be used to identify depression. However, few studies have attempted a large-scale differential diagnosis of patients with depressive disorders using acoustic characteristics of non-English speakers. Objective: This study proposes a framework for automatic depression detection using large-scale acoustic characteristics based on the Korean language. Methods: We recruited 153 patients who met the criteria for major depressive disorder and 165 healthy controls without current or past mental illness. Participants' voices were recorded on a smartphone while performing the task of reading predefined text-based sentences. Three approaches were evaluated and compared to detect depression using data sets with text-dependent read speech tasks: conventional machine learning models based on acoustic features, a proposed model that trains and classifies log-Mel spectrograms by applying a deep convolutional neural network (CNN) with a relatively small number of parameters, and models that train and classify log-Mel spectrograms by applying well-known pretrained networks. Results: The acoustic characteristics of the predefined text-based sentence reading automatically detected depression using the proposed CNN model. The highest accuracy achieved with the proposed CNN on the speech data was 78.14%. Our results show that the deep-learned acoustic characteristics lead to better performance than those obtained using the conventional approach and pretrained models. Conclusions: Checking the mood of patients with major depressive disorder and detecting the consistency of objective descriptions are very important research topics. This study suggests that the analysis of speech data recorded while reading text-dependent sentences could help predict depression status automatically by capturing the characteristics of depression. Our method is smartphone based, is easily accessible, and can contribute to the automatic identification of depressive states.

We present an optimal method for decoding the activity of primary motor cortex (M1) neurons in a nonhuman primate during single finger movements. The method is based on the maximum-likelihood (ML) inference, which assuming the probability of finger movements is uniform, is equivalent to the maximum a posteriori (MAP) inference. Each neuron's activation is first quantified by the change in firing rate before and after finger movement. We then estimate the probability density function of this activation given finger movement, i.e., Pr(neuronal activation (x)| finger movements (m)). Based on the ML criterion, we choose finger movements to maximize Pr(x|m). Experimentally, data were collected from 115 task-related neurons in M1 as the monkey performed flexion and extension of each finger and the wrist (12 movements). With as few as 20-25 randomly selected neurons, the proposed method decoded single-finger movements with 99% accuracy. Since the training and decoding procedures in the proposed method are simple and computationally efficient, the method can be extended for real-time neuroprosthetic control of a dexterous hand.

Frequency modulated continuous wave (FMCW) radar, which can detect the range and small displacement of a target, has been used for contactless vital information extraction. For accurate vital sign (respiration and heartbeat) measurement, a precise selection of the target range bin, where vital information exists, is important. In this paper, an effective method for selecting the range bin with accurate vital information is proposed. The proposed method is based on the newly introduced magnitude-phase coherency (MPC) index. The experimental results show that the vital information extracted by the proposed method is more accurate than those by conventional methods, indicating that the proposed range bin selection based on MPC is an effective method for extracting accurate respiration and heartbeat rates.

You KJ, et al. Exp Neurobiol. 2010 Jun;19(1):54-61. http://dx.doi.org/10.5607/en.2010.19.1.54

In this paper, we provide the results of multi-passenger occupancy detection inside a vehicle obtained using a single-channel frequency-modulated continuous-wave radar. The physiological characteristics of the radar signal are analyzed in a time-frequency spectrum, and features are proposed based on these characteristics for multi-passenger occupancy detection. After clutter removal is applied, the spectral power and Wiener entropy are proposed as features to quantify physiological movements arising from breathing and heartbeat. Using the average means of both the power and Wiener entropy at seats 1 and 2, the feature distributions are expressed, and classification is performed. The multi-passenger occupancy detection performance is evaluated using linear discriminant analysis and maximum likelihood estimation. The results indicate that the proposed power and Wiener entropy are effective features for multi-passenger occupancy detection.

In order to establish a reliable map of the road environment, this paper aims to classify the stationary and moving objects unlike the previous researches which generally focus on object recognition. The characteristics of the radar signals of stationary and moving objects were analyzed and the relation between the slope of pattern in radar time-frequency spectrum and relative velocity of the object was described mathematically. To discriminate the stationary and moving objects, the difference between the measured velocity by the slope and the velocity of the ego-vehicle was proposed as a feature. The statistical characteristics of stationary and moving objects according to the proposed feature were modeled using Gaussian model. To investigate the performance of the proposed method, the similarity between modeling of stationary and moving objects was quantified. Additionally, the receiver operating characteristics (ROC) curve and the correlation coefficient between the proposed feature and the ground-truth feature map was applied to verify the performance.