3D CNN Research Articles

Multicenter Integration of MR Radiomics, Deep Learning, and Clinical Indicators for Predicting Hepatocellular Carcinoma Recurrence After Thermal Ablation.

To develop and validate an innovative predictive model that integrates multisequence magnetic resonance (MR) radiomics, deep learning features, and clinical indicators to accurately predict the recurrence of hepatocellular carcinoma (HCC) after thermal ablation. This retrospective multicenter cohort study enrolled patients who were diagnosed with HCC and treated via thermal ablation. We extracted radiomic features from multisequence 3T MR images, analyzed these images using a 3D convolutional neural network (3D CNN), and incorporated clinical data into the model. Model performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. The study included 535 patients from three hospitals, comprising 462 males and 43 females. The RDC model, which stands for the Radiomics-Deep Learning-Clinical data model, demonstrated high predictive accuracy, achieving AUCs of 0.794 in the training set, 0.777 in the validation set, and 0.787 in the test set. Statistical analysis confirmed the model's robustness and the significant contribution of the integrated features to its predictive capabilities. The RDC model effectively predicts HCC recurrence after thermal ablation by synergistically combining advanced imaging analysis and clinical parameters. This study highlights the potential of such integrative approaches to enhance prognostic assessments in HCC patients and offers a promising tool for clinical decision-making.

Trajectory representation learning with multilevel attention for driver identification

Massive trajectory data have originated from the development of positioning technology. Learning GPS trajectory representation to characterize a driver’s driving style is a challenging task with important applications in many areas, including autonomous driving, auto insurance, advanced driver assistance systems, urban computing, and the internet of things. Few studies have considered the interactions between different factors. In this study, we propose a novel trajectory representation method based on a multilevel attention mechanism (ATTraj2vec) and apply it to the task of driver identification. We use 1D CNN to summarize local features, including motion, spatial and temporal features. Then we utilize a multilevel attention mechanism to extract global features, aggregating the interactions of motion features with temporal and spatial features progressively. Additionally, we adopt multi-loss to optimize our model simultaneously, which consists of a softmax loss for driver classification and Siamese loss for making trajectories from the same driver more similar. Classification experimental results on a real-world automobile trajectory dataset demonstrated that our proposed model significantly outperforms existing baselines. Meanwhile, the proposed method provides significant gains in the trajectory clustering of unseen drivers. The code is available in the repository, https://github.com/limengyuan2021/ATTraj2vec,

Research on fault diagnosis and feature extraction mechanism visualization of rotating machinery based on improved 1D CNN

With the wide application of intelligent equipment in modern industrial production, it is particularly important to study how to intelligently sense the faults of rotating machinery, improve the diagnosis efficiency and enhance the interpretation ability of the diagnosis process. Although the traditional 1-D CNN performs well in fault diagnosis, it has limitations in capturing subtle changes and complex patterns of fault signals, and its interpretability needs to be improved. Therefore, based on the improved ELCNN model, this paper discusses its diagnosis mechanism in depth, aiming at providing a new idea for intelligent fault diagnosis of rotating machinery. The functions of convolutional layer and S-GAP layer in ELCNN are studied and analyzed. Through single-layer linear convolution, ELCNN can adaptively learn the frequency domain features of the signal and realize the lightweight of the model. At the same time, the S-GAP layer enhances the ability of ELCNN to capture the main peak frequency of fault signals through feature sparseness. The experimental results show that the accuracy of ELCNN in frequency domain feature extraction is more than 80 %. The main peak frequency extracted can effectively help engineers understand the basis of model judgment and improve the reliability of the model.

MULTI-LEVEL CLASSIFICATION BASED ON DEEP LEARNING FOR ACCURATE RISK STRATIFICATION OF ARRHYTHMIAS

Arrhythmias, also known as irregular heartbeats, are important health problems that must be accurately identified to diagnose and treat cardiovascular disease. Within the scope of this study, a network for classifying arrhythmias, which are important in the diagnosis and treatment of cardiovascular diseases, was proposed by using one-dimensional convolutional neural network (1D CNN), one of the deep learning techniques. With the proposed 1D-CNN architecture, arrhythmia types and normal rhythm ECGs were subjected to a more detailed examination from general to specific according to urgency situations. In the classifications made, first of all, a binary classification was made and an evaluation was made as whether there was a life risk or not. In triple, quadruple and six-fold classification, the detection of arrhythmia status is detailed. More complex classifications have helped to define different types of arrhythmias in more detail. This study proposes a deep learning network for automatic identification and classification of arrhythmias and shows that different arrhythmia conditions can be diagnosed with a single network model by applying the proposed network structure to multi-class arrhythmia disorders.

Ultra-low cost and high-performance paper-based flexible pressure sensor for artificial intelligent E-skin

E-skin driven by artificial intelligence algorithms enables the innovation of future technologies such as IoT, intelligent manufacturing, health monitoring and human–machine interfaces for VR/AR. Although many flexible pressure sensors have been reported, the development of sensors that combine low-cost and high performance remains a huge challenge. Herein, we propose an all-paper-based pressure sensor with water-soluble graphite functional layer and carbon interdigital electrodes obtained by friction coating and screen printing, respectively. The sensor boasts a rectangular design measuring 1 cm × 2.15 cm, with a remarkably thin profile of only 71 μm and an ultra-light weight of 6 mg, and the cost of a single sensor is extremely low. Specifically, the sensor achieved a lower detection limit of 0.98 Pa and an upper detection limit of 2000 kPa, which represents a new benchmark among paper-based pressure sensors. It demonstrated high sensitivity across various pressure ranges: 319.94 kPa−1 (0–1 kPa), 97.63 kPa−1 (1–40 kPa), 18.75 kPa−1 (40–200 kPa), and 2.72 kPa−1 (200–2000 kPa). The sensor also exhibited a rapid response and recovery time of 8 ms and 9 ms, respectively, along with excellent durability (>20,000 fatigue tests, 10,000 cycles at 2 kPa and 10,000 cycles at 100 kPa) and high pressure stability. Furthermore, the eco-friendly sensor can identify eight surface textures with a recognition accuracy of 98.33% after training with a 2D convolutional neural network (2D CNN). Finally, a portable wireless wearable device with dimensions of 38 mm × 26 mm × 14 mm and a weight of only 11.265 g has been developed for pressure recognition, breathing, pulse, and heartbeat detection.

Turbocharging protein binding site prediction with geometric attention, inter-resolution transfer learning, and homology-based augmentation

BackgroundLocating small molecule binding sites in target proteins, in the resolution of either pocket or residue, is critical in many drug-discovery scenarios. Since it is not always easy to find such binding sites using conventional methods, different deep learning methods to predict binding sites out of protein structures have been developed in recent years. The existing deep learning based methods have several limitations, including (1) the inefficiency of the CNN-only architecture, (2) loss of information due to excessive post-processing, and (3) the under-utilization of available data sources.MethodsWe present a new model architecture and training method that resolves the aforementioned problems. First, by layering geometric self-attention units on top of residue-level 3D CNN outputs, our model overcomes the problems of CNN-only architectures. Second, by configuring the fundamental units of computation as residues and pockets instead of voxels, our method reduced the information loss from post-processing. Lastly, by employing inter-resolution transfer learning and homology-based augmentation, our method maximizes the utilization of available data sources to a significant extent.ResultsThe proposed method significantly outperformed all state-of-the-art baselines regarding both resolutions—pocket and residue. An ablation study demonstrated the indispensability of our proposed architecture, as well as transfer learning and homology-based augmentation, for achieving optimal performance. We further scrutinized our model’s performance through a case study involving human serum albumin, which demonstrated our model’s superior capability in identifying multiple binding sites of the protein, outperforming the existing methods.ConclusionsWe believe that our contribution to the literature is twofold. Firstly, we introduce a novel computational method for binding site prediction with practical applications, substantiated by its strong performance across diverse benchmarks and case studies. Secondly, the innovative aspects in our method— specifically, the design of the model architecture, inter-resolution transfer learning, and homology-based augmentation—would serve as useful components for future work.

Unconscious classification of quantitative electroencephalogram features from propofol versus propofol combined with etomidate anesthesia using one-dimensional convolutional neural network.

Establishing a convolutional neural network model for the recognition of characteristic raw electroencephalogram (EEG) signals is crucial for monitoring consciousness levels and guiding anesthetic drug administration. This trial was conducted from December 2023 to March 2024. A total of 40 surgery patients were randomly divided into either a propofol group (1% propofol injection, 10 mL: 100 mg) (P group) or a propofol-etomidate combination group (1% propofol injection, 10 mL: 100 mg, and 0.2% etomidate injection, 10 mL: 20 mg, mixed at a 2:1 volume ratio) (EP group). In the P group, target-controlled infusion (TCI) was employed for sedation induction, with an initial effect site concentration set at 5-6 μg/mL. The EP group received an intravenous push with a dosage of 0.2 mL/kg. Six consciousness-related EEG features were extracted from both groups and analyzed using four prediction models: support vector machine (SVM), Gaussian Naive Bayes (GNB), artificial neural network (ANN), and one-dimensional convolutional neural network (1D CNN). The performance of the models was evaluated based on accuracy, precision, recall, and F1-score. The power spectral density (94%) and alpha/beta ratio (72%) demonstrated higher accuracy as indicators for assessing consciousness. The classification accuracy of the 1D CNN model for anesthesia-induced unconsciousness (97%) surpassed that of the SVM (83%), GNB (81%), and ANN (83%) models, with a significance level of p < 0.05. Furthermore, the mean and mean difference ± standard error of the primary power values for the EP and P groups during the induced period were as follows: delta (23.85 and 16.79, 7.055 ± 0.817, p < 0.001), theta (10.74 and 8.743, 1.995 ± 0.7045, p < 0.02), and total power (24.31 and 19.72, 4.588 ± 0.7107, p < 0.001). Large slow-wave oscillations, power spectral density, and the alpha/beta ratio are effective indicators of changes in consciousness during intravenous anesthesia with a propofol-etomidate combination. These indicators can aid anesthesiologists in evaluating the depth of anesthesia and adjusting dosages accordingly. The 1D CNN model, which incorporates consciousness-related EEG features, represents a promising tool for assessing the depth of anesthesia. https://www.chictr.org.cn/index.html.

Indirect prediction of the 3D printability of polysaccharide gels using multiple machine learning (ML) models

In this paper, the capabilities of NIR spectroscopy and LF-NMR data were compared for rapidly predicting the rheological properties of polysaccharide gels and assessing their printability. Seven machine learning (ML) models were established for rheological property prediction based on partial least squares regression (PLSR), support vector regression (SVR), back propagation artificial neural network (BPANN), one-dimensional convolutional neural network (1D CNN), recurrent neural network (RNN), long short-term memory (LSTM), and Transformer. The results showed that among the seven models, the SVR, BPANN, and 1D CNN models based on NIR spectroscopy effectively predicted the rheological parameters of polysaccharide gels, with the highest R2 in the prediction set reaching 0.9796 and the highest RPD reaching 7.0708. For most polysaccharide gels, using the LF-NMR relaxation time distribution curves provided better predictions of rheological properties than using transverse relaxation time and peak area. Among the seven models, the PLSR, SVR, 1D CNN, and Transformer models effectively predicted the rheological characteristics based on LF-NMR parameters, with the highest R2 in the prediction set reaching 0.9869 and the highest RPD reaching 8.7220. This study successfully established a prediction system for the rheological behaviors and 3D printing performance of polysaccharide gels using NIR spectroscopy and LF-NMR data combined with ML methods, achieving an intelligent assessment of the 3D printing behavior of polysaccharide gels.

EEG spatial projection and an improved 3D CNN with channel spatiotemporal joint attention mechanism for emotion recognition

EEG spatial projection and an improved 3D CNN with channel spatiotemporal joint attention mechanism for emotion recognition

Artificial Intelligence and Machine Learning in Healthcare Cybersecurity of Current Applications and Future Directions

Data communication has increased due to the growth of technology-driven services and application deployment. Because there are always security risks to networks, the increasing throughput of networks has also increased the likelihood of cybercrimes. Additionally, the abrupt end of the COVID crisis, which caused nearly all services to go online, has increased awareness of the threat posed by cyberattacks in general. Thus, in the information and technology-driven world of today, reliable communication is crucial. The network is frequently discovered to be vulnerable to a variety of network attacks, such as spoofing, malware, and DDoS attacks. The difficulties caused by intrusions have also been covered in the current work, along with the methods to prevent and detect these kinds of intrusion attempts that have been under investigation for the past few years. Research needs have been identified by presenting and analysing the body of literature currently available on IDS. A hybrid approach has been suggested based on the information to provide a secure communication mechanism and protect the network from intruders. A cyber security strategy based on ABC and 3D CNN architecture is suggested as a solution to this. The qualities that reflect malware and DDoS attack aspects are chosen and optimised using the ABC. Neural networks are trained and classified using the retrieved data. The accuracy, recall, and f-measure analysis of the suggested cyber security method for malware and DDoS node detection were assessed. Comparative investigation demonstrated that the cyber security method was more effective at identifying network attacks.

Designing a TPMS metamaterial via deep learning and topology optimization

Data-driven models that act as surrogates for computationally costly 3D topology optimization techniques are very popular because they help alleviate multiple time-consuming 3D finite element analyses during optimization. In this study, one such 3D CNN-based surrogate model for the topology optimization of Schoen’s gyroid triply periodic minimal surface unit cell is investigated. Gyroid-like unit cells are designed using a voxel algorithm and homogenization-based topology optimization codes in MATLAB. A few such optimization data are used as input–output for supervised learning of the topology-optimization process via the 3D CNN model in Python code. These models could then be used to instantaneously predict the optimized unit cell geometry for any topology parameters. The high accuracy of the model was demonstrated by a low mean square error metric and a high Dice coefficient metric. The model has the major disadvantage of running numerous costly topology optimization runs but has the advantages that the trained model can be reused for different cases of TO and that the methodology of the accelerated design of 3D metamaterials can be extended for designing any complex, computationally costly problems of metamaterials with multi-objective properties or multiscale applications. The main purpose of this paper is to provide the complete associated MATLAB and PYTHON codes for optimizing the topology of any cellular structure and predicting new topologies using deep learning for educational purposes.

Optimizing protein content prediction in rice bean (Vigna umbellata L.) by integrating near-infrared reflectance spectroscopy, MPLS, deep learning, and key wavelengths selection algorithms

Rice bean, a potential protein-rich underutilized crop, can serve as a valuable source for food and feed, contributing to sustainable food and nutrition security. However, traditional methods for measuring protein content are laborious, tedious, and expensive. This study evaluates the performance of Modified Partial Least Squares (MPLS) and 1D Convolutional Neural Network (1D CNN) models in predicting protein content in rice bean using Near-Infrared Reflectance Spectroscopy (NIRS) data. Both models were validated with whole spectra, key wavelengths, and spectral segments derived from Genetic Algorithm (GA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Competitive Adaptive Reweighted Sampling (CARS). The 1D CNN model demonstrated superior predictive performance, with an 11 % increase in R² and an 18 % increase in RPD using the whole spectra compared to the MPLS model. Specifically, the 1D CNN model with CARS-selected wavelengths achieved an R² of 0.82 and an RPD of 2.15, while the MPLS model achieved an R² of 0.84 and an RPD of 2.25. The segmented spectra approach further enhanced the 1D CNN model's performance, achieving the highest R² of 0.84 and an RPD of 2.28 with CARS-selected segments. These results highlight that the 1D CNN model with CARS-selected wavelengths provides the best predictive performance for protein content in rice bean. The optimized method can lead to rapid early-phase breeding practices and improved quality assessments in leguminous crops, emphasizing the importance of integrating advanced modeling techniques with NIRS for agricultural applications.

A robust adapted Flexible Parallel Neural Network architecture for early prediction of lithium battery lifespan

Early prediction of End of Life (EPEOL) is crucial for improving lithium battery efficiency and lifespan. Traditional fixed-architecture neural networks often suffer from underfitting or overfitting due to diverse data distributions. To address this, we propose the Flexible Parallel Neural Network (FPNN). This model integrates modules like InceptionBlock, 3D CNN, 2D CNN, and dual-stream networks. By effectively extracting electrochemical features from video-format data through the 3D CNN and achieving multi-scale feature abstraction with InceptionBlock, FPNN ensures effective module coordination. The model can adaptively adjust the number of InceptionBlocks to handle tasks of varying complexity. Experimental results on the MIT dataset show that FPNN achieves MAPE values of 1.26%, 0.41%, 0.37%, 0.33%, 0.32%, 0.32%, 0.31%, 0.31%, 0.22%, and 0.34% using the first 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 cycle data, respectively. The interpretability of FPNN is reflected in its structural design and flexible unit choices, providing a basis for model interpretation. Comprising multiple modules, FPNN enables smooth feature extraction from electrochemical data through collaborative interaction. The diverse branching structure of the flexible units allows the model to capture features at different scales, learning richer information. Our approach offers a precise, adaptable, and easy-to-understand solution for EPEOL, opening new possibilities in battery health monitoring.

Enhancing acceleration data cleaning through time–frequency feature integration using deep learning

In recent years, the safety issues of bridge has garnered significant attention from various sectors of society. Bridge structural health monitoring (SHM) systems have emerged as essential tools for real-time monitoring and assessing the operational safety of bridges. Despite the substantial implementation and operation of monitoring systems, challenges have arisen from abnormal data caused by equipment malfunctions, power outages, and environmental interference. These anomalies are challenging to manually remove within the vast dataset. To address the challenge of automated detection and classification of anomalies, this paper focuses on acceleration data and proposes a data cleaning method that integrates time–frequency features and one-dimensional convolutional neural network (1D CNN). The method utilizes extreme value envelopes and wavelet scattering transforms to extract time–frequency features from acceleration data, thereby achieving data dimensionality reduction and feature extraction goals. Subsequently, these features are employed to train and test a 1D CNN. The proposed method’s validation is conducted using acceleration monitoring data obtained from a large-span arch bridge. Through extensive testing, the model demonstrates a classification accuracy exceeding 98% on both the test set and unseen data, showcasing high classification precision and generalization capabilities. This approach provides an effective solution for automated data cleaning and anomaly identification within the context of bridge health monitoring systems.

A hybrid surrogate model for real-time coastal urban flood prediction: An application to Macao

Tropical cyclones could cause co-occurrences of storm tides and rainfall and, therefore, induce flooding hazards, posing high risks to coastal cities. Accurate and timely prediction of flood hazards in time–space domain is essential for flood risk management. In this study, a hybrid surrogate model is proposed for real-time flooding prediction by considering the compound effects of storm tides, rainfall, and drainage outflows. The hybrid surrogate model combines a long short-term memory (LSTM) neural network for predicting drainage outflows with a one-dimensional convolutional neural network (1D CNN) for predicting water depths. The model is tested using performance metrics, and the simulation results agree well with the historical measurements. Moreover, a sensitivity analysis of the drainage outflows using the modified Morris screening method reveals the importance of considering drainage systems in flood modelling, particularly when rainfall return periods are relatively small. The sensitivity indices vary over the computation domain in terms of different driving factors, with positive values corresponding to rainfall-dominated area while negative values corresponding to storm tide-dominated area. The computation cost of the model is quite low, and the fast prediction speed could provide a sound basis for early warning and real-time flood management. The proposed hybrid surrogate model offers a promising function for rapid and accurate flood inundation prediction in coastal cities, enabling more effective flood risk management and emergency response planning.

Classification of Alzheimer’s dementia EEG signals using deep learning

Alzheimer’s dementia (AD) is a predominant neurological disorder arising from corruptions in brain functions and is characterized by a chronic or progressive nature. While the precise etiology of dementia remains incompletely elucidated, its manifestation is frequently associated with discernible structural and chemical alterations in the brain. Living with dementia significantly impacts individuals’ daily lives due to the resultant loss of cognitive functions. This study presents a novel method to monitor and detect AD using advanced signal processing applied to electroencephalography (EEG) signals. The intrinsic time-scale decomposition (ITD) algorithm is employed to extract proper rotation components (PRCs) from EEG signals, utilizing a 5-second EEG segment duration. The proposed method is compared with the detection of 5-second raw EEG segments using a custom one-dimensional convolutional neural network (1D CNN). Additionally, four different quartiles (Quartile 1 (Q1), Q2, Q3, and Q4) of EEG signals are considered to identify the most significant contributor to AD. Experimental results demonstrate that the ITD-based approach yields better detection performance compared to using raw EEG signals. The most promising result is achieved by the EEG-PRCs method in Q1, with an accuracy of 94.00%, sensitivity of 93.50%, and specificity of 93.90%. In contrast, the highest-performing result of the raw EEG segments method is in Q2, with an accuracy of 88.40%, sensitivity of 89.10%, and specificity of 87.60% in terms of detecting AD.

Human Violence Detection in Videos Using Key Frame Identification and 3D CNN with Convolutional Block Attention Module

Human Violence Detection in Videos Using Key Frame Identification and 3D CNN with Convolutional Block Attention Module

Complex-Valued 2D-3D Hybrid Convolutional Neural Network with Attention Mechanism for PolSAR Image Classification

The recently introduced complex-valued convolutional neural network (CV-CNN) has shown considerable advancements for polarimetric synthetic aperture radar (PolSAR) image classification by effectively incorporating both magnitude and phase information. However, a solitary 2D or 3D CNN encounters challenges such as insufficiently extracting scattering channel dimension features or excessive computational parameters. Moreover, these networks’ default is that all information is equally important, consuming vast resources for processing useless information. To address these issues, this study presents a new hybrid CV-CNN with the attention mechanism (CV-2D/3D-CNN-AM) to classify PolSAR ground objects, possessing both excellent computational efficiency and feature extraction capability. In the proposed framework, multi-level discriminative features are extracted from preprocessed data through hybrid networks in the complex domain, along with a special attention block to filter the feature importance from both spatial and channel dimensions. Experimental results performed on three PolSAR datasets demonstrate our present approach’s superiority over other existing ones. Furthermore, ablation experiments confirm the validity of each module, highlighting our model’s robustness and effectiveness.

Virtual Restoration of Ancient Mold-Damaged Painting Based on 3D Convolutional Neural Network for Hyperspectral Image

Painted cultural relics hold significant historical value and are crucial in transmitting human culture. However, mold is a common issue for paper or silk-based relics, which not only affects their preservation and longevity but also conceals the texture, patterns, and color information, hindering cultural value and heritage. Currently, the virtual restoration of painting relics primarily involves filling in the RGB based on neighborhood information, which might cause color distortion and other problems. Another approach considers mold as noise and employs maximum noise separation for its removal; however, eliminating the mold components and implementing the inverse transformation often leads to more loss of information. To effectively acquire virtual restoration for mold removal from ancient paintings, the spectral characteristics of mold were analyzed. Based on the spectral features of mold and the cultural relic restoration philosophy of maintaining originality, a 3D CNN artifact restoration network was proposed. This network is capable of learning features in the near-infrared spectrum (NIR) and spatial dimensions to reconstruct the reflectance of visible spectrum, achieving the virtual restoration for mold removal of calligraphic and art relics. Using an ancient painting from the Qing Dynasty as a test subject, the proposed method was compared with the Inpainting, Criminisi, and inverse MNF transformation methods across three regions. Visual analysis, quantitative evaluation (the root mean squared error (RMSE), mean absolute percentage error (MAPE), mean absolute error (MEA), and a classification application were used to assess the restoration accuracy. The visual results and quantitative analyses demonstrated that the proposed 3D CNN method effectively removes or mitigates mold while restoring the artwork to its authentic color in various backgrounds. Furthermore, the color classification results indicated that the images restored with 3D CNN had the highest classification accuracy, with overall accuracies of 89.51%, 92.24%, and 93.63%, and Kappa coefficients of 0.88, 0.91, and 0.93, respectively. This research provides technological support for the digitalization and restoration of cultural artifacts, thereby contributing to the preservation and transmission of cultural heritage.

Neural Decoding of Spontaneous Overt and Intended Speech.

The aim of this study was to decode intended and overt speech from neuromagnetic signals while the participants performed spontaneous overt speech tasks without cues or prompts (stimuli). Magnetoencephalography (MEG), a noninvasive neuroimaging technique, was used to collect neural signals from seven healthy adult English speakers performing spontaneous, overt speech tasks. The participants randomly spoke the words yes or no at a self-paced rate without cues. Two machine learning models, namely, linear discriminant analysis (LDA) and one-dimensional convolutional neural network (1D CNN), were employed to classify the two words from the recorded MEG signals. LDA and 1D CNN achieved average decoding accuracies of 79.02% and 90.40%, respectively, in decoding overt speech, significantly surpassing the chance level (50%). The accuracy for decoding intended speech was 67.19% using 1D CNN. This study showcases the possibility of decoding spontaneous overt and intended speech directly from neural signals in the absence of perceptual interference. We believe that these findings make a steady step toward the future spontaneous speech-based brain-computer interface.