The existing automatic sleep staging algorithms have the problems of too many model parameters and long training time, which in turn results in poor sleep staging efficiency. Using a single channel electroencephalogram (EEG) signal, this paper proposed an automatic sleep staging algorithm for stochastic depth residual networks based on transfer learning (TL-SDResNet). Firstly, a total of 30 single-channel (Fpz-Cz) EEG signals from 16 individuals were selected, and after preserving the effective sleep segments, the raw EEG signals were pre-processed using Butterworth filter and continuous wavelet transform to obtain two-dimensional images containing its time-frequency joint features as the input data for the staging model. Then, a ResNet50 pre-trained model trained on a publicly available dataset, the sleep database extension stored in European data format (Sleep-EDFx) was constructed, using a stochastic depth strategy and modifying the output layer to optimize the model structure. Finally, transfer learning was applied to the human sleep process throughout the night. The algorithm in this paper achieved a model staging accuracy of 87.95% after conducting several experiments. Experiments show that TL-SDResNet50 can accomplish fast training of a small amount of EEG data, and the overall effect is better than other staging algorithms and classical algorithms in recent years, which has certain practical value.
Humans ; Sleep Stages ; Algorithms ; Sleep ; Wavelet Analysis ; Electroencephalography/methods* ; Machine Learning

OBJECTIVE: To propose a semi-supervised material quantitative intelligent imaging algorithm based on prior information perception learning (SLMD-Net) to improve the quality and precision of spectral CT imaging. METHODS: The algorithm includes a supervised and a self- supervised submodule. In the supervised submodule, the mapping relationship between low and high signal-to-noise ratio (SNR) data was constructed through mean square error loss function learning based on a small labeled dataset. In the self- supervised sub-module, an image recovery model was utilized to construct the loss function incorporating the prior information from a large unlabeled low SNR basic material image dataset, and the total variation (TV) model was used to to characterize the prior information of the images. The two submodules were combined to form the SLMD-Net method, and pre-clinical simulation data were used to validate the feasibility and effectiveness of the algorithm. RESULTS: Compared with the traditional model-driven quantitative imaging methods (FBP-DI, PWLS-PCG, and E3DTV), data-driven supervised-learning-based quantitative imaging methods (SUMD-Net and BFCNN), a material quantitative imaging method based on unsupervised learning (UNTV-Net) and semi-supervised learning-based cycle consistent generative adversarial network (Semi-CycleGAN), the proposed SLMD-Net method had better performance in both visual and quantitative assessments. For quantitative imaging of water and bone materials, the SLMD-Net method had the highest PSNR index (31.82 and 29.06), the highest FSIM index (0.95 and 0.90), and the lowest RMSE index (0.03 and 0.02), respectively) and achieved significantly higher image quality scores than the other 7 material decomposition methods (P < 0.05). The material quantitative imaging performance of SLMD-Net was close to that of the supervised network SUMD-Net trained with labeled data with a doubled size. CONCLUSIONS A small labeled dataset and a large unlabeled low SNR material image dataset can be fully used to suppress noise amplification and artifacts in basic material decomposition in spectral CT and reduce the dependence on labeled data-driven network, which considers more realistic scenario in clinics.
Tomography, X-Ray Computed/methods* ; Image Processing, Computer-Assisted/methods* ; Algorithms ; Signal-To-Noise Ratio ; Perception

Due to the high complexity and subject variability of motor imagery electroencephalogram, its decoding is limited by the inadequate accuracy of traditional recognition models. To resolve this problem, a recognition model for motor imagery electroencephalogram based on flicker noise spectrum (FNS) and weighted filter bank common spatial pattern ( wFBCSP) was proposed. First, the FNS method was used to analyze the motor imagery electroencephalogram. Using the second derivative moment as structure function, the ensued precursor time series were generated by using a sliding window strategy, so that hidden dynamic information of transition phase could be captured. Then, based on the characteristic of signal frequency band, the feature of the transition phase precursor time series and reaction phase series were extracted by wFBCSP, generating features representing relevant transition and reaction phase. To make the selected features adapt to subject variability and realize better generalization, algorithm of minimum redundancy maximum relevance was further used to select features. Finally, support vector machine as the classifier was used for the classification. In the motor imagery electroencephalogram recognition, the method proposed in this study yielded an average accuracy of 86.34%, which is higher than the comparison methods. Thus, our proposed method provides a new idea for decoding motor imagery electroencephalogram.
Brain-Computer Interfaces ; Imagination ; Signal Processing, Computer-Assisted ; Electroencephalography/methods* ; Algorithms ; Spectrum Analysis

Heart valve disease (HVD) is one of the common cardiovascular diseases. Heart sound is an important physiological signal for diagnosing HVDs. This paper proposed a model based on combination of basic component features and envelope autocorrelation features to detect early HVDs. Initially, heart sound signals lasting 5 minutes were denoised by empirical mode decomposition (EMD) algorithm and segmented. Then the basic component features and envelope autocorrelation features of heart sound segments were extracted to construct heart sound feature set. Then the max-relevance and min-redundancy (MRMR) algorithm was utilized to select the optimal mixed feature subset. Finally, decision tree, support vector machine (SVM) and k-nearest neighbor (KNN) classifiers were trained to detect the early HVDs from the normal heart sounds and obtained the best accuracy of 99.9% in clinical database. Normal valve, abnormal semilunar valve and abnormal atrioventricular valve heart sounds were classified and the best accuracy was 99.8%. Moreover, normal valve, single-valve abnormal and multi-valve abnormal heart sounds were classified and the best accuracy was 98.2%. In public database, this method also obtained the good overall accuracy. The result demonstrated this proposed method had important value for the clinical diagnosis of early HVDs.
Humans ; Heart Sounds ; Heart Valve Diseases/diagnosis* ; Algorithms ; Support Vector Machine ; Signal Processing, Computer-Assisted

Anti-motion artifact is one of the most important properties of ambulatory ECG monitoring equipment. At present, there is a lack of standardized means to test the performance of anti-motion artifact. ECG simulator and special conductive leather are used to build the simulator, it is used to simulate human skin, to generate ECG signal input for the ECG monitoring equipment attached to it. The mechanical arm and fixed support are used to build a motion simulation system to fix the conductive leather. The mechanical arm is programmed to simulate various motion states of the human body, so that the ECG monitoring equipment can produce corresponding motion artifacts. The collected ECG signals are read wirelessly, observed, analyzed and compared, and the anti-motion artifact performance of ECG monitoring equipment is evaluated. The test results show that by artificially creating the small difference between the two groups of ambulatory ECG monitoring equipment, the system can accurately test the interference signals introduced under the conditions of controlled movement such as tension and torsion, and compare the advantages and disadvantages. The research shows that the test system can provide convenient and accurate verification means for the research of optimizing anti-motion interference.
Humans ; Artifacts ; Signal Processing, Computer-Assisted ; Electrocardiography, Ambulatory/methods* ; Electrocardiography ; Motion

The acoustic detection method based on machine learning and signal processing is an important method of pathological voice detection and the extraction of voice features is one of the most important. Currently, the features widely used have disadvantage of dependence on the fundamental frequency extraction, being easily affected by noise and high computational complexity. In view of these shortcomings, a new method of pathological voice detection based on multi-band analysis and chaotic analysis is proposed. The gammatone filter bank was used to simulate the human ear auditory characteristics to analyze different frequency bands and obtain the signals in different frequency bands. According to the characteristics that turbulence noise caused by chaos in voice will worsen the spectrum convergence, we applied short time Fourier transform to each frequency band of the voice signal, then the feature gammatone short time spectral self-similarity (GSTS) was extracted, and the chaos degree of each band signal was analyzed to distinguish normal and pathological voice. The experimental results showed that combined with traditional machine learning methods, GSTS reached the accuracy of 99.50% in the pathological voice database of Massachusetts Eye and Ear Infirmary (MEEI) and had an improvement of 3.46% compared with the best existing features. Also, the time of the extraction of GSTS was far less than that of traditional nonlinear features. These results show that GSTS has higher extraction efficiency and better recognition effect than the existing features.
Acoustics ; Databases, Factual ; Fourier Analysis ; Humans ; Noise ; Signal Processing, Computer-Assisted

High resolution (HR) magnetic resonance images (MRI) or computed tomography (CT) images can provide clearer anatomical details of human body, which facilitates early diagnosis of the diseases. However, due to the imaging system, imaging environment and human factors, it is difficult to obtain clear high-resolution images. In this paper, we proposed a novel medical image super resolution (SR) reconstruction method via multi-scale information distillation (MSID) network in the non-subsampled shearlet transform (NSST) domain, namely NSST-MSID network. We first proposed a MSID network that mainly consisted of a series of stacked MSID blocks to fully exploit features from images and effectively restore the low resolution (LR) images to HR images. In addition, most previous methods predict the HR images in the spatial domain, producing over-smoothed outputs while losing texture details. Thus, we viewed the medical image SR task as the prediction of NSST coefficients, which make further MSID network keep richer structure details than that in spatial domain. Finally, the experimental results on our constructed medical image datasets demonstrated that the proposed method was capable of obtaining better peak signal to noise ratio (PSNR), structural similarity (SSIM) and root mean square error (RMSE) values and keeping global topological structure and local texture detail better than other outstanding methods, which achieves good medical image reconstruction effect.
Humans ; Image Processing, Computer-Assisted/methods* ; Signal-To-Noise Ratio ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed

Bowel sounds is an important indicator to monitor and reflect intestinal motor function, and traditional manual auscultation requires high professional knowledge and rich clinical experience of doctors. In addition, long-time auscultation is time-consuming and laborious， which may lead to misjudgment caused by subjective error. To solve the problem, firstly, the wavelet transform is used to preprocess the bowel sounds signal for noise reduction and enhancement. Secondly, three typical features of intestinal sound were extracted. According to the combination of these features, a three-stage decision was designed to carry out multi-parameter and multi-feature joint threshold detection. This algorithm realized the detection of bowel sound signal and the location of its start and end points, making it possible that the complete bowel sound signal was extracted effectively. In this study, a large number of clinical data and label of bowel sounds were collected, and a new effective evaluation method was proposed to verify the proposed method. The accuracy rate is 83.51%. Results of this study will provide systematic support and theoretical guarantee for the diagnosis of intestinal diseases and the monitoring of postoperative intestinal function recovery of patients.
Algorithms ; Auscultation ; Humans ; Intestines ; Signal Processing, Computer-Assisted ; Wavelet Analysis

OBJECTIVE: Non-contact continuous blood pressure monitoring is significant in vital sign monitoring. Frequency modulated continuous wave (FMCW) radar is suitable for non-contact wave signal extraction. A heartbeat-guided blood pressure monitoring algorithm using FMCW radar is proposed. METHODS: The target heart rate is detected and pulse wave signal is extracted based on FMCW reflected signals. The variational mode decomposition (VMD) is introduced to alleviate the interferences of human breath and slight body movements. The pulse wave signal is extracted based on target heart rate. Blood pressure related features of pure pulse waveform are extracted to obtain blood pressure based on its estimation model. RESULTS: Experiments are conducted indoor among 15 participants sitting in a nature state. The average accuracy of diastolic blood pressure (DBP) is 94.3% and that of systolic blood pressure is 94.4%. CONCLUSIONS The experimental results demonstrate the robustness and effectiveness of the proposed algorithm, which makes it possible to further achieve long-term real-time non-contact blood pressure monitoring.
Algorithms ; Blood Pressure ; Blood Pressure Determination ; Heart Rate ; Humans ; Radar ; Signal Processing, Computer-Assisted ; Vital Signs

Photoplethysmography (PPG) is a non-invasive technique to measure heart rate at a lower cost, and it has been recently widely used in smart wearable devices. However, as PPG is easily affected by noises under high-intensity movement, the measured heart rate in sports has low precision. To tackle the problem, this paper proposed a heart rate extraction algorithm based on self-adaptive heart rate separation model. The algorithm firstly preprocessed acceleration and PPG signals, from which cadence and heart rate history were extracted respectively. A self-adaptive model was made based on the connection between the extracted information and current heart rate, and to output possible domain of the heart rate accordingly. The algorithm proposed in this article removed the interference from strong noises by narrowing the domain of real heart rate. From experimental results on the PPG dataset used in 2015 IEEE Signal Processing Cup, the average absolute error on 12 training sets was 1.12 beat per minute (bpm) (Pearson correlation coefficient: 0.996; consistency error: -0.184 bpm). The average absolute error on 10 testing sets was 3.19 bpm (Pearson correlation coefficient: 0.990; consistency error: 1.327 bpm). From experimental results, the algorithm proposed in this paper can effectively extract heart rate information under noises and has the potential to be put in usage in smart wearable devices.
Algorithms ; Heart Rate/physiology* ; Photoplethysmography/methods* ; Signal Processing, Computer-Assisted ; Wearable Electronic Devices