Search

Abstract Electroencephalography (EEG) measures changes in neuronal activity and can reveal significant changes from infancy to adulthood concomitant with brain maturation, making it a potential physiological marker of brain maturation and cognition. To investigate a promising deep learning tool for EEG classification, we applied the bidirectional long short-term memory (BLSTM) algorithm to analyze EEG data from the pediatric EEG laboratory of Taipei Tzu Chi Hospital. The trained BLSTM model was 86% accurate when identifying EEGs from young children (8 months–6 years) and adolescents (12–20 years). However, there was only a modest classification accuracy (69.3%) when categorizing EEG samples into three age groups (8 months–6 years, 6–12 years, and 12–20 years). For EEG samples from patients with intellectual disability, the prediction accuracy of the trained BLSTM model was 46.4%, which was significantly lower than its accuracy for EEGs from neurotypical patients, indicating that the individual’s intelligence plays a major role in the age prediction. This study confirmed that scalp EEG can reflect brain maturation and the BLSTM algorithm is a feasible deep learning tool for the identification of cognitive age. The trained model can potentially be applied to clinical services as a supportive measurement of neurodevelopmental status.

IntroductionAttention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder affecting children worldwide; however, diagnosing ADHD remains a complex task. Theta/beta ratio (TBR) derived from electroencephalography (EEG) recordings has been proposed as a potential biomarker for ADHD, but its effectiveness in children with ADHD remains controversial. Behavioral assessments, such as the Conners Continuous Performance Test–3rd edition (CPT-3), have been utilized to assess attentional capacity in individuals with ADHD. This study aims to investigate the correlation between TBR and CPT-3 scores in children and adolescents with ADHD.MethodsIn a retrospective analysis, we examined patients regularly monitored for ADHD at Taipei Tzu Chi Hospital, who underwent both EEG and CPT-3 assessments. Severity of ADHD was evaluated using parent- and teacher-completed Swanson, Nolan, and Pelham (SNAP)-IV rating scales.ResultsThe study encompassed 55 ADHD patients (41 with abnormal CPT-3 scores, 14 with normal CPT-3 scores) and 45 control subjects. TBR demonstrated elevation in ADHD patients with abnormal CPT-3 scores, indicating its potential to represent attentional capacity akin to behavioral assessments like CPT-3. However, significant correlations between TBR values and CPT-3 variables or SNAP-IV rating scales were not observed. Moreover, TBR values exhibited considerable overlap across the groups, leading to diminished sensitivity and negative predictive value as a potential neurophysiological ADHD biomarker.DiscussionWhile our study underscores the utility of both TBR and CPT-3 in assessing attentional capacity, their sensitivity in diagnosing ADHD is limited. A comprehensive evaluation, integrating clinical expertise, parental input, and detailed neuropsychometric tests, remains pivotal for a thorough and precise diagnosis of ADHD.

Goal: Numerous studies had successfully differentiated normal and abnormal voice samples. Nevertheless, further classification had rarely been attempted. This study proposes a novel approach, using continuous Mandarin speech instead of a single vowel, to classify four common voice disorders (i.e. functional dysphonia, neoplasm, phonotrauma, and vocal palsy). Methods: In the proposed framework, acoustic signals are transformed into mel-frequency cepstral coefficients, and a bi-directional long-short term memory network (BiLSTM) is adopted to model the sequential features. The experiments were conducted on a large-scale database, wherein 1,045 continuous speech were collected by the speech clinic of a hospital from 2012 to 2019. Results: Experimental results demonstrated that the proposed framework yields significant accuracy and unweighted average recall improvements of 78.12–89.27% and 50.92–80.68%, respectively, compared with systems that use a single vowel. Conclusions: The results are consistent with other machine learning algorithms, including gated recurrent units, random forest, deep neural networks, and LSTM.The sensitivities for each disorder were also analyzed, and the model capabilities were visualized via principal component analysis. An alternative experiment based on a balanced dataset again confirms the advantages of using continuous speech for learning voice disorders.

This study presents a deep learning-based speech signal-processing mobile application known as CITISEN. The CITISEN can perform three functions: speech enhancement (SE), model adaptation (MA), and background noise conversion (BNC), which allow CITISEN to be used as a platform for utilizing and evaluating SE models and flexibly extend the models to address various noise environments and users. For SE, CITISEN downloads pretrained SE models on the cloud server and then uses these models to effectively reduce noise components from prerecordings or instant recordings provided by users. When it encounters noisy speech signals with unknown speakers or noise types, the MA function allows CITISEN to improve the SE performance effectively. A few audio files of unseen speakers or noise types are recorded and uploaded to the cloud server and then used to adapt the pretrained SE model. Finally, for BNC, CITISEN removes the original background noise using an SE model and then mixes the processed speech signal with new background noise. The novel BNC function can evaluate SE performance under specific conditions, cover people’s tracks, and provide entertainment. The experimental results confirmed the effectiveness of SE, MA, and BNC functions. Compared with the noisy speech signals, the enhanced speech signals by SE achieved about 6% and 33% of improvements, respectively, in terms of short-time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ). With MA, the STOI and PESQ could be further improved by approximately 6% and 11%, respectively. Note that the SE model and MA method are not limited to the ones described in this study and can be replaced with any SE model and MA method. Finally, the BNC experiment results indicated that the speech signals of original and converted backgrounds have a close scene identification accuracy and similar embeddings in an acoustic scene classification model. Therefore, the proposed BNC can effectively convert the background noise of a speech signal and be a data augmentation method when clean speech signals are unavailable.

The detection of audio tampering plays a crucial role in ensuring the authenticity and integrity of multimedia files. This paper presents a novel approach to identifying tampered audio files by leveraging the unique Electric Network Frequency (ENF) signal, which is inherent to the power grid and serves as a reliable indicator of authenticity. The study begins by establishing a comprehensive Chinese ENF database containing diverse ENF signals extracted from audio files. The proposed methodology involves extracting the ENF signal, applying wavelet decomposition, and utilizing the autoregressive model to train effective classification models. Subsequently, the framework is employed to detect audio tampering and assess the influence of various environmental conditions and recording devices on the ENF signal. Experimental evaluations conducted on our Chinese ENF database demonstrate the efficacy of the proposed method, achieving impressive accuracy rates ranging from 91% to 93%. The results emphasize the significance of ENF-based approaches in enhancing audio file forensics and reaffirm the necessity of adopting reliable tamper detection techniques in multimedia authentication.

Introduction: Attention problems are frequently observed in patients with Prader–Willi syndrome (PWS); however, only few studies have investigated the severity and mechanisms of attention problems in them. In this study, we aim to evaluate dynamic changes in the quantitative electroencephalographic (EEG) spectrum during attention tasks in patients with PWS.Method: From January to June 2019, 10 patients with PWS and 10 age-matched neurotypical control participants were recruited at Taipei Tzu Chi Hospital. Each participant completed Conners’ continuous performance test, third edition (CPT-3), tasks with simultaneous EEG monitoring. The dynamic changes in the quantitative EEG spectrum between the resting state and during CPT-3 tasks were compared.Results: Behaviorally, patients with PWS experienced significant attention problems, indicated by the high scores for several CPT-3 variables. The theta/beta ratio of the resting-state EEG spectrum revealed no significant differences between the control participants and patients with PWS. During CPT-3 tasks, a significant decrease in the alpha power was noted in controls compared with that in patients with PWS. The attention-to-resting alpha power ratio was positively correlated with many CPT-3 variables. After adjusting for genotype, age, intelligence, and body mass index, the attention-to-resting alpha power ratio was still significantly correlated with participants’ commission errors.Conclusion: This study provides evidence that attention problems are frequently observed in patients with PWS, while attention impairment can be demonstrated by dynamic changes in the quantitative EEG spectrum.

Deep learning techniques such as convolutional neural networks (CNN) have been successfully applied to identify pathological voices. However, the major disadvantage of using these advanced models is the lack of interpretability in explaining the predicted outcomes. This drawback further introduces a bottleneck for promoting the classification or detection of voice-disorder systems, especially in this pandemic period. In this paper, we proposed using a series of learnable sinc functions to replace the very first layer of a commonly used CNN to develop an explainable SincNet system for classifying or detecting pathological voices. The applied sinc filters, a front-end signal processor in SincNet, are critical for constructing the meaningful layer and are directly used to extract the acoustic features for following networks to generate high-level voice information. We conducted our tests on three different Far Eastern Memorial Hospital voice datasets. From our evaluations, the proposed approach achieves the highest 7%–accuracy and 9%–sensitivity improvements from conventional methods and thus demonstrates superior performance in predicting input pathological waveforms of the SincNet system. More importantly, we intended to give possible explanations between the system output and the first-layer extracted speech features based on our evaluated results.

Nonnegative matrix factorization (NMF) is a useful decomposition technique for multivariate data. More recently, NMF technology was used as a noise-estimation stage for a Wiener-filtering-based noise reduction (NR) method to improve the quality of noisy speech. Previous studies showed that this method provides better sound quality performance than conventional NMF-based approaches; however, there is still scope for improving the performance under noisy listening conditions. More specifically, the performance of an NMF noise estimator for calculating the noise level is considered sensitive to diverse noise environments and signal-to-noise ratio conditions. Therefore, we proposed an adaptive algorithm that derives an adaptive factor ($\alpha $ ) to adjust the weight between the estimated speech and noise levels on the basis of the signal-to-noise level for the gain function of the Wiener-filtering-based NR method to further improve the sound quality. Two objective evaluations and listening tests evaluated the benefits of the proposed method, and experimental results show that better output sound quality and competed for speech intelligibility performance can be achieved when compared with conventional unsupervised NR and NMF-based methods.

In wireless telephony and audio data mining applications, it is desirable that noise suppression can be made robust against changing noise conditions and operates in real time (or faster). The learning effectiveness and speed of artificial neural networks are therefore critical factors in applications for speech enhancement tasks. To address these issues, we present an extreme learning machine (ELM) framework, aimed at the effective and fast removal of background noise from a single-channel speech signal, based on a set of randomly chosen hidden units and analytically determined output weights. Because feature learning with shallow ELM may not be effective for natural signals, such as speech, even with a large number of hidden nodes, hierarchical ELM (H-ELM) architectures are deployed by leveraging sparse autoencoders. In this manner, we not only keep all the advantages of deep models in approximating complicated functions and maintaining strong regression capabilities, but we also overcome the cumbersome and time-consuming features of both greedy layer-wise pre-training and back-propagation (BP)-based fine tuning schemes, which are typically adopted for training deep neural architectures. The proposed ELM framework was evaluated on the Aurora-4 speech database. The Aurora-4 task provides relatively limited training data, and test speech data corrupted with both additive noise and convolutive distortions for matched and mismatched channels and signal-to-noise ratio (SNR) conditions. In addition, the task includes a subset of testing data involving noise types and SNR levels that are not seen in the training data. The experimental results indicate that when the amount of training data is limited, both ELMand H-ELM-based speech enhancement techniques consistently outperform the conventional BP-based shallow and deep learning algorithms, in terms of standardized objective evaluations, under various testing conditions.

Conventionally, the maximum likelihood (ML) criterion is applied to train a deep belief network (DBN). We present a maximum entropy (ME) learning algorithm for DBNs, designed specifically to handle limited training data. Maximizing only the entropy of parameters in the DBN allows more effective generalization capability, less bias towards data distributions, and robustness to over-fitting compared to ML learning. Results of text classification and object recognition tasks demonstrate ME-trained DBN outperforms ML-trained DBN when training data is limited.