Speech emotion recognition via graph-based representations.

Speech emotion recognition (SER) has gained an increased interest during the last decades as part of enriched affective computing. As a consequence, a variety of engineering approaches have been developed addressing the challenge of the SER problem, exploiting different features, learning algorithms, and datasets. In this paper, we propose the application of the graph theory for classifying emotionally-colored speech signals. Graph theory provides tools for extracting statistical as well as structural information from any time series. We propose to use the mentioned information as a novel feature set. Furthermore, we suggest setting a unique feature-based identity for each emotion belonging to each speaker. The emotion classification is performed by a Random Forest classifier in a Leave-One-Speaker-Out Cross Validation (LOSO-CV) scheme. The proposed method is compared with two state-of-the-art approaches involving well known hand-crafted features as well as deep learning architectures operating on mel-spectrograms. Experimental results on three datasets, EMODB (German, acted) and AESDD (Greek, acted), and DEMoS (Italian, in-the-wild), reveal that our proposed method outperforms the comparative methods in these datasets. Specifically, we observe an average UAR increase of almost [Formula: see text], [Formula: see text] and [Formula: see text], respectively.

Using the recurrence plots as indicators for the recognition of Parkinson's disease through phonemes assessment.

Parkinson's disease (PD) is considered to be the second most common neurodegenerative disease which affects the patients' life throughout the years. As a consequence, its early diagnosis is of major importance for the improvement of life quality, implying that the severe symptoms can be delayed through appropriate clinical intervention and treatment. Among the most important premature symptoms of PD are the voice impairments of articulation, phonation and prosody. The objective of this study is to investigate whether the voice's dynamic behavior can be used as possible indicator for PD. Thus in this work, we employ the recurrence plots (RPs) which derive from the analysis of the three modulated vowels /a/, /e/ and /o/, which belong to the PC-GITA dataset, and are fed as input images to a 3-channel Convolutional Neural Network-based (CNN) architecture, which, finally, differentiates the 50 PD patients from 50 healthy subjects. The experimental results obtained provide evidence that the RP-based approach is a promising tool for the recognition of PD patients through the analysis of voice recordings, with a classification accuracy achieved equal to 87%.

Altered hippocampal connectivity dynamics predicts memory performance in neuropsychiatric lupus: a resting-state fMRI study using cross-recurrence quantification analysis.

OBJECTIVE: Τo determine whole-brain and regional functional connectivity (FC) characteristics of patients with neuropsychiatric SLE (NPSLE) or without neuropsychiatric manifestations (non-NPSLE) and examine their association with cognitive performance. METHODS: Cross-recurrence quantification analysis (CRQA) of resting-state functional MRI (rs-fMRI) data was performed in 44 patients with NPSLE, 20 patients without NPSLE and 35 healthy controls (HCs). Volumetric analysis of total brain and specific cortical and subcortical regions, where significant connectivity changes were identified, was performed. Cognitive status of patients with NPSLE was assessed by neuropsychological tests. Group comparisons on nodal FC, global network metrics and regional volumetrics were conducted, and associations with cognitive performance were estimated (at p<0.05 false discovery rate corrected). RESULTS: FC in patients with NPSLE was characterised by increased modularity (mean (SD)=0.31 (0.06)) as compared with HCs (mean (SD)=0.27 (0.06); p=0.05), hypoconnectivity of the left (mean (SD)=0.06 (0.018)) and right hippocampi (mean (SD)=0.051 (0.0.16)), and of the right amygdala (mean (SD)=0.091 (0.039)), as compared with HCs (mean (SD)=0.075 (0.022), p=0.02; 0.065 (0.019), p=0.01; 0.14 (0.096), p=0.05, respectively). Hyperconnectivity of the left angular gyrus (NPSLE/HCs: mean (SD)=0.29 (0.26) and 0.10 (0.09); p=0.01), left (NPSLE/HCs: mean (SD)=0.16 (0.09) and 0.09 (0.05); p=0.01) and right superior parietal lobule (SPL) (NPSLE/HCs: mean (SD)=0.25 (0.19) and 0.13 (0.13), p=0.01) was noted in NPSLE versus HC groups. Among patients with NPSLE, verbal episodic memory scores were positively associated with connectivity (local efficiency) of the left hippocampus (r2=0.22, p=0.005) and negatively with local efficiency of the left angular gyrus (r2=0.24, p=0.003). Patients without NPSLE displayed hypoconnectivity of the right hippocampus (mean (SD)=0.056 (0.014)) and hyperconnectivity of the left angular gyrus (mean (SD)=0.25 (0.13)) and SPL (mean (SD)=0.17 (0.12)). CONCLUSION: By using dynamic CRQA of the rs-fMRI data, distorted FC was found globally, as well as in medial temporal and parietal brain regions in patients with SLE, that correlated significantly and adversely with memory capacity in NPSLE. These results highlight the value of dynamic approaches to assessing impaired brain network function in patients with lupus with and without neuropsychiatric symptoms.

Multi-Modal Deep Learning Diagnosis of Parkinson's Disease-A Systematic Review.

Parkinson's Disease (PD) is among the most frequent neurological disorders. Approaches that employ artificial intelligence and notably deep learning, have been extensively embraced with promising outcomes. This study dispenses an exhaustive review between 2016 and January 2023 on deep learning techniques used in the prognosis and evolution of symptoms and characteristics of the disease based on gait, upper limb movement, speech and facial expression-related information as well as the fusion of more than one of the aforementioned modalities. The search resulted in the selection of 87 original research publications, of which we have summarized the relevant information regarding the utilized learning and development process, demographic information, primary outcomes, and sensory equipment related information. Various deep learning algorithms and frameworks have attained state-of-the-art performance in many PD-related tasks by outperforming conventional machine learning approaches, according to the research reviewed. In the meanwhile, we identify significant drawbacks in the existing research, including a lack of data availability and interpretability of models. The fast advancements in deep learning and the rise in accessible data provide the opportunity to address these difficulties in the near future and for the broad application of this technology in clinical settings.

Classification of Compressed Remote Sensing Multispectral Images via Convolutional Neural Networks.

Multispectral sensors constitute a core Earth observation image technology generating massive high-dimensional observations. To address the communication and storage constraints of remote sensing platforms, lossy data compression becomes necessary, but it unavoidably introduces unwanted artifacts. In this work, we consider the encoding of multispectral observations into high-order tensor structures which can naturally capture multi-dimensional dependencies and correlations, and we propose a resource-efficient compression scheme based on quantized low-rank tensor completion. The proposed method is also applicable to the case of missing observations due to environmental conditions, such as cloud cover. To quantify the performance of compression, we consider both typical image quality metrics as well as the impact on state-of-the-art deep learning-based land-cover classification schemes. Experimental analysis on observations from the ESA Sentinel-2 satellite reveals that even minimal compression can have negative effects on classification performance which can be efficiently addressed by our proposed recovery scheme.

Survey of Deep-Learning Approaches for Remote Sensing Observation Enhancement.

Deep Learning, and Deep Neural Networks in particular, have established themselves as the new norm in signal and data processing, achieving state-of-the-art performance in image, audio, and natural language understanding. In remote sensing, a large body of research has been devoted to the application of deep learning for typical supervised learning tasks such as classification. Less yet equally important effort has also been allocated to addressing the challenges associated with the enhancement of low-quality observations from remote sensing platforms. Addressing such channels is of paramount importance, both in itself, since high-altitude imaging, environmental conditions, and imaging systems trade-offs lead to low-quality observation, as well as to facilitate subsequent analysis, such as classification and detection. In this paper, we provide a comprehensive review of deep-learning methods for the enhancement of remote sensing observations, focusing on critical tasks including single and multi-band super-resolution, denoising, restoration, pan-sharpening, and fusion, among others. In addition to the detailed analysis and comparison of recently presented approaches, different research avenues which could be explored in the future are also discussed.