Concatenated Convolutional Neural Network Model for Cuffless Blood Pressure Estimation Using Fuzzy Recurrence Properties of PPG Signals

Due to the importance of continuous monitoring of blood pressure (BP) in controlling hypertension, the topic of cuffless blood pressure (BP) estimation has been widely studied in recent years. A most important approach is to explore the nonlinear mapping between the recorded peripheral signals and the BP values which is usually conducted by deep neural networks. Because of the sequence-based pseudo periodic nature of peripheral signals such as photoplethysmogram (PPG), a proper estimation model needed to be equipped with the 1-dimensional (1-D) and recurrent layers. This, in turn, limits the usage of 2-dimensional (2-D) layers adopted in convolutional neural networks (CNN) for embedding spatial information in the model. In this study, considering the advantage of chaotic approaches, the recurrence characterization of peripheral signals was taken into account by a visual 2-D representation of PPG in phase space through fuzzy recurrence plot (FRP). FRP not only provides a beneficial framework for capturing the spatial properties of input signals but also creates a reliable approach for embedding the pseudo periodic properties to the neural models without using recurrent layers. Moreover, this study proposes a novel deep neural network architecture that combines the morphological features extracted simultaneously from two upgraded 1-D and 2-D CNNs capturing the temporal and spatial dependencies of PPGs in systolic and diastolic BP estimation. The model has been fed with the 1-D PPG sequences and the corresponding 2-D FRPs from two separate routes. The performance of the proposed framework was examined on the well-known public dataset, namely, Multi-Parameter Intelligent in Intensive Care II. Our scheme is analyzed and compared with the literature in terms of the requirements of the standards set by the British Hypertension Society (BHS) and the Association for the Advancement of Medical Instrumentation (AAMI). The proposed model met the AAMI requirements, and it achieved a grade of A as stated by the BHS standard. In addition, its mean absolute errors (MAE) and standard deviation for both systolic and diastolic blood pressure estimations were considerably low, 3.05±5.26 mmHg and 1.58±2.6 mmHg, in turn.

Human Muscle Synergy Analysis to Approach the Understanding of Brain Control Algorithm

Purpose: Muscle synergy is a functional unit that coordinates the activity of a number of muscles. In this study, the extraction of muscle synergies in three types of hand movements in the horizontal plane is investigated. Materials and Methods: So, after constructing the tracking pattern of three signals, by LabVIEW, the Electromyography (EMG) signal from six muscles of hand was recorded. Then time-constant muscle synergies and their activity curves from the recorded EMG signals were extracted using Non-negative Matrix Factorization (NMF) method. Results: Comparison of these patterns showed that the non-random motions’ synergies were more similar than the random motions among different individuals. It was observed that in all movements, the similarity of the synergies in one cluster was greater than the similarity of their corresponding activation curves. Conclusion: The results showed that the complexity of the recurrence plot in random movement is greater than that of the other movements.

Alternate Entropy Computations by Applying Recurrence Matrix Masking

In practicality, recurrence analyses of dynamical systems can only process short sections of signals that may be infinitely long. By necessity, the recurrence plot and its quantifications are constrained within a truncated triangle that clips the signals at its borders. Recurrence variables defined within these confining borders can be influenced more or less by truncation effects depending upon the system under evaluation. In this study, the question being asked is what if the boundary borders were tilted, what would be the effect on all recurrence variables? This question was prompted by the observation that line entropy values are maximized for highly periodic systems in which the infinitely long line elements are truncated to different unique lengths. However, by redefining the recurrence plot area to a 45-degree tilted box within the triangular area, the diagonal lines would consequently be truncated to identical lengths. Such masking would minimize the line entropy to 0.000 bits/bin. However, what new truncation influences would be imposed on the other recurrence variables? This question is examined by comparing recurrence variables computed with the triangular recurrence area versus boxed recurrence area. Examples include the logistic equation (mathematical series), the Dow Jones Industrial Average over a decade (real-word data), and a square wave pulse (toy series). Good agreement among the variables in terms of timing and amplitude was found for most, but not all variables. These important results are discussed.

Recurrence quantification analysis of Q&A behavior from the perspective of explicit and tacit knowledge – an empirical study based on Zhihu's hashtags

PurposeThe purpose of this research is to investigate the time structure characteristics of collaborative knowledge production behaviors in Q&A (question-and-answer) communities for explicit and tacit knowledge, and systematically investigate the supply side and the demand side of knowledge production.Design/methodology/approachTaking Zhihu as the research object, using the methods of recurrence plot and recurrence quantification analysis, this paper analyzes the recursive characteristics of the motion trajectories of the three behavioral sequences of questioning, answering, and discussion, qualitatively and quantitatively analyzing the generation and evolution mechanism of explicit and tacit knowledge.FindingsThe results show that compared with the demand-side behavior sequence, the supply-side behavior sequence exhibits higher stability, complexity and periodicity. Compared with the tacit knowledge topics, the demand-side behavior sequence of the explicit knowledge topics shows stronger nonlinearity, and the supply-side behavior sequence shows lower complexity.Originality/valueThe research conclusions provide preliminary evidence for the effectiveness of the recurrence plot method in distinguishing different types of knowledge production behaviors and have important application value for the “crowdsourcing” knowledge generation and identification under the knowledge economy and the sustainable development of the socialized question-and-answer community.

Classification of Chaotic Squeak and Rattle Vibrations by CNN Using Recurrence Pattern

The chaotic squeak and rattle (S&R) vibrations in mechanical systems were classified by deep learning. The rattle, single-mode, and multi-mode squeak models were constructed to generate chaotic S&R signals. The repetition of nonlinear signals generated by them was visualized using an unthresholded recurrence plot and learned using a convolutional neural network (CNN). The results showed that even if the signal of the S&R model is chaos, it could be classified. The accuracy of the classification was verified by calculating the Lyapunov exponent of the vibration signal. The numerical experiment confirmed that the CNN classification using nonlinear vibration images as the proposed procedure has more than 90% accuracy. The chaotic status and each model can be classified into six classes.

Dynamic Analysis of Blast Furnace Sensor Data using Cross-recurrence Quantification Strategies

This paper investigates impact degree of blast furnace related elements towards blast furnace gas (BFG) production. BFG is a by-product in the steel industry, which is one of the enterprise’s most essential energy resources. While because multiple factors affect BFG production it has characteristics of large fluctuations. Most works focus on finding a satisfactory method or improving the accuracy of existing methods to predict BFG production. There are no special studies on the factors that affect the production of BFG. Finding the elements that affect BFG production is benefit to production of BFG, which has a significance in economy. We propose a novel framework, combining cross recurrence plot (CRP) and cross recurrence quantification analysis (CRQA). Moreover, it supplies a general method to convert time series of BFG related data into high-dimensional space. This is the first analytical framework that attempts to reveal the inherent dynamic similarities of blast furnace gas-related elements. The experimental results demonstrate that this framework can realize the visualization of the time series. In addition, the results also identify the factor that has the greatest impact on blast furnace gas production by quantitative analysis.

Time-Lapse Image Method for Classifying Appliances in Nonintrusive Load Monitoring

In this paper, a time-lapse image method is proposed to improve the classification accuracy for multistate appliances with complex patterns based on nonintrusive load monitoring (NILM). A log-likelihood ratio detector with a maxima algorithm was applied to construct a real-time event detection of home appliances. Moreover, a novel image-combining method was employed to extract information from the data based on the Gramian angular field (GAF) and recurrence plot (RP) transformations. From the simulation results, it was confirmed that the classification accuracy can be increased by up to 30% with the proposed method compared with the conventional approaches in classifying multistate appliances.