Schizophrenia is a severe mental disease that affects patients' thoughts, feelings, and behaviors. Speech signal has proven to be a biomarker in the early diagnosis of schizophrenia. Previous studies on schizophrenic speech detection are mainly based on manual feature extraction engineering, which requires domain knowledge for researchers and has difficulties extracting effective features. This work proposes an end-to-end architecture, called Axial-attention-based Network using Wideband and Narrowband Spectrograms (WNSA-Net), to detect schizophrenia. Specifically, we adopt both wideband and narrowband spectrograms as inputs to represent speech signals using fine time and frequency structures. Then dilated convolution blocks are employed to capture detailed and long-range information in spectrograms. Axial-attention blocks are introduced to augment the information in feature maps along the time and frequency axes. In addition, we employ a gate mechanism to fuse the output feature maps from all channels. Experimental results on the Schizophrenia dataset and its subdatasets show that schizophrenic patients have difficulties in expressing emotions. To validate the performance of our WNSA-Net, experiments are conducted on Schizophrenia dataset and open-access TORGO database, achieving 97.37% and 98.16% accuracy in detecting schizophrenia and dysarthria, respectively. The results show promise for the proposed method in the diagnosis of disordered speech.

Location-based social networks (LBSNs) add geographical information into traditional social networks and link people’s virtual and physical lives. As an important application of LBSNs, point-of-interest (POI) suggestion has become an important method to help users explore interesting and attractive locations in LBSNs. The main problems of POI suggestion include data sparsity and cold start, which have been paid much attention by existing techniques. There are two major challenges which can greatly influence the performance of suggestion accuracy. One is the fuzzy boundary between sentiments, i.e., the fine distinction between sentiments makes it difficult to classify words and texts after word-sentiment mapping operation. The other challenge is the unreliability of data quality represented by similarity metrics, which relies on data integrity and path reachability of a heterogeneous network. To cope with the above two challenges, we present a novel framework calledCommunity-based SentimentExtraction andNetworkEmbedding for POIRecommendation (CENTER) for suggesting impressive POIs to a specific user in an effective fashion. The CENTER framework contains two essential techniques: (1) a latent probabilistic generative model calledCommunity-basedSentimentExtraction (CSE), which can accurately capture the sentiments from review content in LBSNs by taking into consideration the characteristics of social communities. The parameters of the CSE model can be inferred effectively by the Gibbs sampling method. The primary sentiments are obtained based on the distribution of sentiments; (2) a network embedding model calledSentiment-awareNeworkEmbedding for POIRecommendation (SNER) is employed to learn the representation of the factors including POIs, users and textual sentiments in a low-dimensional embedding space. The joint training is utilized to alternatively sample all sets of edges in a heterogeneous information network. Extensive experiments were conducted on two large-scale real datasets, in order to evaluate the performance of the proposed CENTER framework. The results demonstrate that CENTER is superior to the state-of-the-art baseline methods in the effectiveness and efficiency of POI suggestion.

Attention deficit hyperactivity disorder (ADHD) is one of the most common childhood mental disorders. Hyperactivity is a typical symptom of ADHD in children. Clinicians diagnose this symptom by evaluating the children's activities based on subjective rating scales and clinical experience. In this work, an objective system is proposed to quantify the movements of children with ADHD automatically. This system presents a new movement detection and quantification method based on depth images. A novel salient object extraction method is proposed to segment body regions. In movement detection, we explore a new local search algorithm to detect any potential motions of children based on three newly designed evaluation metrics. In the movement quantification, two parameters are investigated to quantify the participation degree and the displacements of each body part in the movements. This system is tested by a depth dataset of children with ADHD. The movement detection results of this dataset mainly range from 91.0% to 95.0%. The movement quantification results of children are consistent with the clinical observations. The public MSR Action 3D dataset is tested to validate the performance of this system.

Among the characteristic symptoms of schizophrenia, negative symptoms are a category. The existence of negative symptoms results in a diminution of normal behaviors and functions for schizophrenic patients. In this research, we focus on the affective flattening of negative symptoms to explore potential biomarkers to achieve automatic diagnosis. This work proposes an automatic procedure for detecting the negative symptoms of schizophrenic patients based on speech signal processing. In this procedure, three features are initially proposed: the symmetric spectral difference level (SSDL), quantization error and vector angle (QEVA), and standard dynamic volume value (SDVV). The SSDL feature is designed with the aim of emphasizing spectral differences that contribute to the evaluation of emotional richness. The QEVA is proposed to reflect the variations in tone using one cumulative error indicator and one variation evaluation indicator. The SDVV feature aims to represent the modulation of speech intensity, considering the speaking behavior of schizophrenic patients. Experiments evaluating the discriminative capabilities of the three features are conducted using a speech database collected from 56 participants (28 schizophrenic patients and 28 healthy controls). The classifier employed in these experiments is a simple decision tree. Three other binary classifiers [linear discrimination (LD), logistic regression (LR), and a support vector machine (SVM)] are also tested to compare their performances with the decision tree in this work. Based on the decision tree classifier, the discrimination accuracy levels of schizophrenic patients and control subjects using the SDLL, QEVA, and SDVV features are in the range of 80.5%-83%, 65%-73%, and 87%- 91.5%, respectively. When the three features are combined, the best discrimination accuracy of schizophrenic patients and control subjects is 98.2%, with an area under the receiver operations characteristic curve (AUC) value of 98%. Note to Practitioners-This article is motivated by the problems of the diagnosis of negative symptoms in schizophrenia and the timely monitoring of schizophrenic patients. In clinic, there exists a high patient-to-clinician ratio and the diagnostic result depends on the subjective experience of the clinician. It is necessary to develop an automatic and objective procedure for the diagnosis and monitoring of schizophrenic patients. Speech disorders are among the salient characteristics of negative symptoms. In this work, an automatic procedure for detecting the negative symptoms based on speech signal processing is proposed. This automatic procedure is achieved based on three newly presented acoustic features involving the characteristics of schizophrenic speech, which consider the speaking behavior of schizophrenic patients. This automatic procedure and the information of the three acoustic features may serve as an aid to clinicians and could potentially help them in providing better monitoring of schizophrenic patients.

Affective flattening is a typical negative symptom in schizophrenia that causes a diminution of normal behaviors and functions in schizophrenic patients. In this work, an automatic algorithm of schizophrenia detection is proposed. This algorithm includes three newly proposed features. These features establish a representation for the production and perception of schizophrenic speech, called the K-Sf (kurtosis skewness fraction), SPI (sound perception indicator), and EBMC (enhanced bilateral matching coefficient). The K-Sf is proposed to reflect the overall distribution of speech segments. The SPI evaluates the relation between the speech composition and the sound perception. The EBMC feature is proposed based on the theory of air pressure oscillations which could reflect the details of air modulation in the vocal tract. Experiments evaluating the discriminative capabilities of the three features are conducted using a speech dataset that is collected from 56 participants (28 schizophrenic patients and 28 healthy controls) and an ensemble classifier. Comparative experiments with SVM classifiers are also conducted. The discrimination accuracies of patients and control subjects using the K-Sf, SPI, EBMC, and the ensemble classifier are in the range of 76.8-92.9%, 82.1-92.8%, and 80.4-91.1%, respectively. When the three features are combined, the discrimination results range from 89.3% to 94.6%. The experimental results indicate that the three features have stronger robustness and better discrimination capability than those previous features relating to the detection of flat affect.

The automatic evaluation of CP (Cleft Plate) speech has various clinical applications. In this work, automatic classification of hypernasality levels and detection of consonant omission methods are proposed. Considering that the data collection is a major bottleneck in the field of CP speech signal processing, an extensive CP speech database is used. This database is collected over 10 years by the Hospital of Stomatology, Sichuan University, which has the largest number of CLP (Cleft Lip and Palate) patients in China. The vocabulary used for this database covers all the initial consonants and the most widely used vowels in Mandarin. Based on the production process of hypernasality, a two-step classification algorithm is proposed to identify four levels of hypernasality: normal, mild, moderate and severe. In order to locate consonant omission, an effective classification algorithm is proposed through analyzing the acoustic characteristics of initial consonants and finals. The classification accuracy for four hypernasality levels reaches up to 83% and the correct identification of consonant omission is over 94%.