Towards high-accuracy classifying attention-deficit/hyperactivity disorders using CNN-LSTM model

ADHD/CD-NET: automated EEG-based characterization of ADHD and CD using explainable deep neural network technique

In this study, attention deficit hyperactivity disorder (ADHD), a childhood neurodevelopmental disorder, is being studied alongside its comorbidity, conduct disorder (CD), a behavioral disorder. Because ADHD and CD share commonalities, distinguishing them is difficult, thus increasing the risk of misdiagnosis. It is crucial that these two conditions are not mistakenly identified as the same because the treatment plan varies depending on whether the patient has CD or ADHD. Hence, this study proposes an electroencephalogram (EEG)-based deep learning system known as ADHD/CD-NET that is capable of objectively distinguishing ADHD, ADHD + CD, and CD. The 12-channel EEG signals were first segmented and converted into channel-wise continuous wavelet transform (CWT) correlation matrices. The resulting matrices were then used to train the convolutional neural network (CNN) model, and the model's performance was evaluated using 10-fold cross-validation. Gradient-weighted class activation mapping (Grad-CAM) was also used to provide explanations for the prediction result made by the 'black box' CNN model. Internal private dataset (45 ADHD, 62 ADHD + CD and 16 CD) and external public dataset (61 ADHD and 60 healthy controls) were used to evaluate ADHD/CD-NET. As a result, ADHD/CD-NET achieved classification accuracy, sensitivity, specificity, and precision of 93.70%, 90.83%, 95.35% and 91.85% for the internal evaluation, and 98.19%, 98.36%, 98.03% and 98.06% for the external evaluation. Grad-CAM also identified significant channels that contributed to the diagnosis outcome. Therefore, ADHD/CD-NET can perform temporal localization and choose significant EEG channels for diagnosis, thus providing objective analysis for mental health professionals and clinicians to consider when making a diagnosis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11571-023-10028-2.

Reconstruction of Eriocheir sinensis Protein-Protein Interaction Network Based on DGO-SVM Method

Eriocheir sinensis is an economically important aquatic animal. Its regulatory mechanisms underlying many biological processes are still vague due to the lack of systematic analysis tools. The protein-protein interaction network (PIN) is an important tool for the systematic analysis of regulatory mechanisms. In this work, a novel machine learning method, DGO-SVM, was applied to predict the protein-protein interaction (PPI) in E. sinensis, and its PIN was reconstructed. With the domain, biological process, molecular functions and subcellular locations of proteins as the features, DGO-SVM showed excellent performance in Bombyx mori, humans and five aquatic crustaceans, with 92-96% accuracy. With DGO-SVM, the PIN of E. sinensis was reconstructed, containing 14,703 proteins and 7,243,597 interactions, in which 35,604 interactions were associated with 566 novel proteins mainly involved in the response to exogenous stimuli, cellular macromolecular metabolism and regulation. The DGO-SVM demonstrated that the biological process, molecular functions and subcellular locations of proteins are significant factors for the precise prediction of PPIs. We reconstructed the largest PIN for E. sinensis, which provides a systematic tool for the regulatory mechanism analysis. Furthermore, the novel-protein-related PPIs in the PIN may provide important clues for the mechanism analysis of the underlying specific physiological processes in E. sinensis.

Wearable-Based Integrated System for In-Home Monitoring and Analysis of Nocturnal Enuresis

Nocturnal enuresis (NE) is involuntary bedwetting during sleep, typically appearing in young children. Despite the potential benefits of the long-term home monitoring of NE patients for research and treatment enhancement, this area remains underexplored. To address this, we propose NEcare, an in-home monitoring system that utilizes wearable devices and machine learning techniques. NEcare collects sensor data from an electrocardiogram, body impedance (BI), a three-axis accelerometer, and a three-axis gyroscope to examine bladder volume (BV), heart rate (HR), and periodic limb movements in sleep (PLMS). Additionally, it analyzes the collected NE patient data and supports NE moment estimation using heuristic rules and deep learning techniques. To demonstrate the feasibility of in-home monitoring for NE patients using our wearable system, we used our datasets from 30 in-hospital patients and 4 in-home patients. The results show that NEcare captures expected trends associated with NE occurrences, including BV increase, HR increase, and PLMS appearance. In addition, we studied the machine learning-based NE moment estimation, which could help relieve the burdens of NE patients and their families. Finally, we address the limitations and outline future research directions for the development of wearable systems for NE patients.

Enhancing generalized anxiety disorder diagnosis precision: MSTCNN model utilizing high-frequency EEG signals

Generalized Anxiety Disorder (GAD) is a prevalent mental disorder on the rise in modern society. It is crucial to achieve precise diagnosis of GAD for improving the treatments and averting exacerbation. Although a growing number of researchers beginning to explore the deep learning algorithms for detecting mental disorders, there is a dearth of reports concerning precise GAD diagnosis. This study proposes a multi-scale spatial-temporal local sequential and global parallel convolutional model, named MSTCNN, which designed to achieve highly accurate GAD diagnosis using high-frequency electroencephalogram (EEG) signals. To this end, 10-min resting EEG data were collected from 45 GAD patients and 36 healthy controls (HC). Various frequency bands were extracted from the EEG data as the inputs of the MSTCNN. The results demonstrate that the proposed MSTCNN, combined with the attention mechanism of Squeeze-and-Excitation Networks, achieves outstanding classification performance for GAD detection, with an accuracy of 99.48% within the 4-30 Hz EEG data, which is competitively related to state-of-art methods in terms of GAD classification. Furthermore, our research unveils an intriguing revelation regarding the pivotal role of high-frequency band in GAD diagnosis. As the frequency band increases, diagnostic accuracy improves. Notably, high-frequency EEG data ranging from 10-30 Hz exhibited an accuracy rate of 99.47%, paralleling the performance of the broader 4-30 Hz band. In summary, these findings move a step forward towards the practical application of automatic diagnosis of GAD and provide basic theory and technical support for the development of future clinical diagnosis system.

Aided diagnosis of cervical spondylotic myelopathy using deep learning methods based on electroencephalography

Cervical spondylotic myelopathy (CSM) is the most severe type of cervical spondylosis. It is challenging to achieve early diagnosis with current clinical diagnostic tools. In this paper, we propose an end-to-end deep learning approach for early diagnosis of CSM. Electroencephalography (EEG) experiments were conducted with patients having spinal cord cervical spondylosis and age-matched normal subjects. A Convolutional Neural Network with Long Short-Term Memory Networks (CNN-LSTM) model was employed for the classification of patients versus normal individuals. In contrast, a Convolutional Neural Network with Bidirectional Long Short-Term Memory Networks and attention mechanism (CNN-BiLSTM-attention) model was used to classify regular, mild, and severe patients. The models were trained using focal Loss instead of traditional cross-entropy Loss, and cross-validation was performed. Our method achieved a classification accuracy of 92.5 % for the two-class classification among 40 subjects and 72.2 % for the three-class classification among 36 subjects. Furthermore, we observed that the proposed model outperformed traditional EEG decoding models. This paper presents an effective computer-aided diagnosis method that eliminates the need for manual extraction of EEG features and holds potential for future auxiliary diagnosis of spinal cord-type cervical spondylosis.

An efficient deep learning framework for P300 evoked related potential detection in EEG signal

BACKGROUND

Incorporating the time-frequency localization properties of Gabor transform (GT), the complexity understandings of convolutional neural network (CNN), and histogram of oriented gradients (HOG) efficacy in distinguishing positive peaks can exhibit their characteristics to reveal an effective solution in the detection of P300 evoked related potential (ERP). By applying a drastic number of convolutional layers, the majority of deep networks elicit sufficient properties for the output determination, leading to gigantic and time-consuming structures. In this paper, we propose a novel deep learning framework by the combination of tuned GT, and modified HOG with the CNN as "TGT-MHOG-CNN" for detection of P300 ERP in EEG signal.

METHOD

In the proposed method, GT is tuned based on triangular function for EEG signals, and then spectrograms including time-frequency information are captured. The function's parameters are justified to differentiate the signals with the P300 component. Furthermore, HOG is modified (MHOG) for the 2-D EEG signal, and consequently, gradients patterns are extracted for the target potentials. MHOG is potent in distinguishing the positive peak in the general waveform; however, GT unravels time-frequency information, which is ignored in the gradient histogram. These outputs of GT and MHOG do not share the same nature in the images nor overlap. Therefore, more extensive information is reached without redundancy or excessive information by fusing them. Combining GT and MHOG provides different patterns which benefit CNN for more precise detection. Consequently, TGT-MHOG-CNN ends in a more straightforward structure than other networks, and therefore, the whole performance is acceptable with faster rates and very high accuracy.

RESULTS

BCI Competition II and III datasets are used to evaluate the performance of the proposed method. These datasets include a complete record for P300 ERP with BCI2000 using a paradigm, and it has numerous noises, including power and muscle-based noises. The objective is to predict the correct character in each provided character selection epochs. Compared to state-of-the-art methods, simulation results indicate striking abilities of the proposed framework for P300 ERP detection. Our best record reached the P300 ERP classification rates of over 98.7% accuracy and 98.7% precision for BCI Competition II and 99% accuracy and 100% precision for BCI Competition III datasets, with superiority in execution time for the mentioned datasets.