The Experts below are selected from a list of 168 Experts worldwide ranked by ideXlab platform
David Zhang - One of the best experts on this subject based on the ideXlab platform.
-
radial artery Pulse waveform analysis based on curve fitting using discrete fourier series
Computer Methods and Programs in Biomedicine, 2019Co-Authors: Zhixing Jiang, David Zhang, Guangming LuAbstract:Abstract Background and objectives: Radial artery Pulse Diagnosis has been playing an important role in traditional Chinese medicine (TCM). For its non-invasion and convenience, the Pulse Diagnosis has great significance in diseases analysis of modern medicine. The practitioners sense the Pulse waveforms in patients’ wrist to make diagnoses based on their non-objective personal experience. With the researches of Pulse acquisition platforms and computerized analysis methods, the objective study on Pulse Diagnosis can help the TCM to keep up with the development of modern medicine. Methods: In this paper, we propose a new method to extract feature from Pulse waveform based on discrete Fourier series (DFS). It regards the waveform as one kind of signal that consists of a series of sub-components represented by sine and cosine (SC) signals with different frequencies and amplitudes. After the Pulse signals are collected and preprocessed, we fit the average waveform for each sample using discrete Fourier series by least squares. The feature vector is comprised by the coefficients of discrete Fourier series function. Results: Compared with the fitting method using Gaussian mixture function, the fitting errors of proposed method are smaller, which indicate that our method can represent the original signal better. The classification performance of proposed feature is superior to the other features extracted from waveform, liking auto-regression model and Gaussian mixture model. Conclusions: The coefficients of optimized DFS function, who is used to fit the arterial pressure waveforms, can obtain better performance in modeling the waveforms and holds more potential information for distinguishing different psychological states.
-
detection of saturation and artifact
2018Co-Authors: David Zhang, Peng WangAbstract:During the Pulse signal acquisition, corruptions would be inevitably introduced such as high-frequency noise, baseline drift, saturation, and artifact. Some of the corrupted Pulse signals can be recovered via preprocessing, but several types of corrupted Pulse signals would be difficult to recover and should be removed from the Pulse signal dataset. Therefore, low-quality Pulse signal detection plays an important role in computational Pulse Diagnosis especially in the real-time Pulse monitoring. In this work, we focus on the detection of two common Pulse corruption types, i.e., saturation and artifact. For the detection of saturation, we use two criteria from its definition. For the artifact detection, we transform the Pulse signal into a complex network and detect the artifact by measuring the connectivity of the network. The experimental results show that the saturation and artifact detection method can both achieve better detection accuracy and better time resolution.
-
introduction computational Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:Pulse Diagnosis is a traditional Diagnosis technique by analyzing the tactile radial arterial palpation by trained fingertips; however it is a subjective skill which needs years of training and practice to master. Computational Pulse Diagnosis intends to employ some modern sensor and computer technology to make Pulse Diagnosis more objective. In this chapter, we will give an overview of computational Pulse Diagnosis. Firstly, the principle of Pulse Diagnosis and the traditional Pulse Diagnosis were introduced, and then the main concept of and the four stages of computational Pulse Diagnosis were introduced.
-
characterization of inter cycle variations for wrist Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:Although Pulse signal is quasiperiodic, most feature extraction methods usually consider it as a whole or only use a single cycle, neglecting the variations between Pulse cycles. To characterize both the inter- and intra-cycle variations, in this chapter we propose three feature extraction methods, i.e., simple combination, multi-scale entropy, and complex network. The simple combination method is a direct extension of conventional single-cycle feature extraction method by concatenating features from multiple cycles. The multi-scale entropy method measures the inter- and intra-cycle variations using entropies of different scales. The complex network method transforms the Pulse signal from time domain to network domain and measures the inter-cycle variations using the statistical properties on complex network. Experimental results show that the presented features are effective in characterizing both inter- and intra-cycle variations and can obtain better performance in Pulse Diagnosis.
-
edit distance for Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:In this chapter, by referring to the edit distance with real penalty (ERP) and the recent progress in k-nearest neighbors (KNN) classifiers, we propose two novel ERP-based KNN classifiers. Taking advantage of the metric property of ERP, we first develop an ERP-induced inner product and a Gaussian ERP kernel, then embed them into difference-weighted KNN classifiers, and finally develop two novel classifiers for Pulse waveform classification. The experimental results show that the proposed classifiers are effective for accurate classification of Pulse waveform.
Yufeng Chung - One of the best experts on this subject based on the ideXlab platform.
-
using an array sensor to determine differences in Pulse Diagnosis three positions and nine indicators
European Journal of Integrative Medicine, 2014Co-Authors: Yuwe Chu, Yufeng Chung, Chinghsing Luo, Chengchang YehAbstract:Abstract Introduction In Pulse Diagnosis, three positions ( Cun , Guan , Chi ) and three levels or depths ( Fu , Zhong , Chen ), called the Three Positions and Nine Indicators (TPNI) are generally used as a finger-reading method, to obtain a holistic view of the patient. However, single Pulse signals at nine TPNI locations (three depths at three positions) look quite similar in a waveform. Hence, the aim of this study was to determine if there was a significant difference between Pulse-taking depths (Fu, Zhong, and Chen) and Pulse-taking positions (Cun, Guan, and Chi). Method To explore the significance of array Pulses at the nine TPNI locations, a Bi-Sensing Pulse Diagnosis Instrument (BSPDI) with array sensors was used to measure wrist artery Pulse signals. It was proposed that a three-dimensional Pulse mapping (3DPM) could present array Pulses mimicking the fingertips’ sensations of a physician. Four parameters, namely peak value ( P _ V max ), power ( P _ P ), ascending slope ( P _AS), and descending slope ( P _DS) were elucidated from 3DPM using a two-way analysis of variance. Eight volunteers with TPNI health rule of thumb from the R.O.C. Air Force Academy participated in this research. Results The variance of four parameters at nine TPNI locations all reached the level of significance ( p Conclusions The differences in wrist artery signals exist between TPNI locations. TPNI Pulse Diagnosis could be used to check the holistic health of a patient as determined by TCM.
-
how to standardize the Pulse taking method of traditional chinese medicine Pulse Diagnosis
Computers in Biology and Medicine, 2013Co-Authors: Yufeng Chung, Chengchang Yeh, Chinghsing LuoAbstract:The aim of this report is to propose standard Pulse taking procedure of Traditional Chinese Medicine Pulse Diagnosis. In order to acquire full information from taking a wrist Pulse, this proposal adopts a tactile sensor with 12 sensing points at one sensing position, such as Cun, Guan, or Chi. Simultaneously Palpation (SP) and Pressing with One Finger (PWOF) are adopted to explore their differences of the detected Pulse signals. According to vertical dynamic characteristics, the results of a Pearson product moment reveal that the correlation coefficients of PWOF and SP are highly correlated from Fu to Chen. In addition, according to unique characteristics of body state, the results of a paired samples t test reveal that the SP and PWOF are indifferent at a specific Pulse taking depth. Hence, if using the Pulse-taking instrument with tactile sensors, it is concluded that Pulse signals taken by familiar SP and PWOF methods are shown in statistical indifferences among seven parameters (V"p"p"m"e"a"n",V"p"p"m"a"x, HR, LENGTH, WIDTH, AS, and DS).
-
possibility of quantifying tcm finger reading sensations i bi sensing Pulse Diagnosis instrument
European Journal of Integrative Medicine, 2012Co-Authors: Chinghsing Luo, Yufeng Chung, Chengchang Yeh, Da Hsua Feng, Yungchu Lee, Shi I Huang, Shu Ming Yeh, Chi Hsie LiangAbstract:Abstract Aim of this study This paper reports the construction and functionality of a newly designed Bi-Sensing Pulse Diagnosis Instrument (BSPDI) with a Pressure-Displacement Bi-Sensing System (PDBSS) coupled to a robot finger system. The BSPDI is used to simulate the three positions and nine indicators (TPNI) using the finger-reading rule, which addresses the three levels of superficial ( Fu ), medium ( Zhong ), and deep ( Chen ) at the three positions of distal ( CUN ), middle ( GUAN ) and proximal ( CHI ) on the wrist during Pulse Diagnosis. Materials and methods A strain gauge and polyvinylidene fluoride pressure sensor were integrated as a PDBSS to separately record finger-reading skills and sense a physician's fingertip sensations. The corresponding sensation and Pulse-taking displacement between the physician's fingertips and the BSPDI and the relationship between the Pulse Diagnosis and the nine TPNI displacements were used to determine the feasibility of the BSPDI. Results TPNI displacements, representing a physician's finger-reading skill, were recorded using a PDBSS. In this scenario, the BSPDI robot fingertips with pressure sensors were placed at the nine recorded TPNI displacements. The pressure sensors recorded Pulse signals to simulate a physician's fingertip sensations. Two groups of participants with different health statuses were evaluated by both the BSPDI and TCM physicians. Both groups received similar diagnoses, and the results of a two-way ANOVA indicated that Pulse diagnoses were highly related to the clinical experiences of the Pulse-taking displacements and positions. Conclusions The novel BSPDI can be used to successfully mimic a TCM physician's finger-reading skill, as this technique obtained Pulse signals at the nine TPNI displacements representing fingertip sensations.
-
temporal and spatial properties of arterial pulsation measurement using pressure sensor array
Evidence-based Complementary and Alternative Medicine, 2012Co-Authors: Chungshing Hu, Yufeng ChungAbstract:Conventionally, a Pulse taking platform is based on a single sensor, which initiates a feasible method of quantitative Pulse Diagnosis. The aim of this paper is to implement a Pulse taking platform with a tactile array sensor. Three-dimensional wrist Pulse signals are constructed, and the length, width, ascending slope, and descending slope are defined following the surface of the wrist Pulse. And the pressure waveform of the wrist Pulse obtained through proposed Pulse-taking platform has the same performance as the single sensor. Finally, the results of a paired samples t-test reveal that the repeatability of the proposal platform is consistent with clinical experience. On the other hand, the results of ANOVA indicate that differences exist among different Pulse taking depths, and this result is consistent with clinical experience in traditional Chinese medicine Pulse Diagnosis (TCMPD). Hence, the proposed Pulse taking platform with an array sensor is feasible for quantification in TCMPD.
-
exploring the conventional Pulse conditions using bi sensing Pulse Diagnosis instrument
BioMedical Engineering and Informatics, 2011Co-Authors: Yufeng Chung, Yuwe Chu, Chinghsing Luo, Chengchang YehAbstract:Pulse Diagnosis is one of the efficient techniques to detect the health status of a patient. However, Pulse Diagnosis suffers from a viewpoint of nonscientific and subjectivity. The goal of this report wants to propose an outline to integrate both the quantification of Pulse Diagnosis and application of clinic. Hence, the definitions of the quantifiable parameter are followed by clinical experiences. These core parameters are classified into four categories: position, rate and rhythm, shape, and trend. Using Bi-Sensing Pulse Diagnosis Instrument (BSPDI) to acquire wrist artery information for interpreting Pulse conditions. The DS (Depth-Strength) curve can be classified the Pulse condition whether it belongs to floating Pulse or sunken Pulse; and similarly, DR (Depth-Rate-Rhythm) curve to rapid Pulse or slow Pulse, DW (Depth-Width) curve to fine Pulse or large Pulse, DL (Depth-Length) curve to long Pulse or short Pulse. The Pulse conditions will be understood based on these curves and Pulse Diagnosis is no more subjectivity and nonscientific.
Peng Wang - One of the best experts on this subject based on the ideXlab platform.
-
detection of saturation and artifact
2018Co-Authors: David Zhang, Peng WangAbstract:During the Pulse signal acquisition, corruptions would be inevitably introduced such as high-frequency noise, baseline drift, saturation, and artifact. Some of the corrupted Pulse signals can be recovered via preprocessing, but several types of corrupted Pulse signals would be difficult to recover and should be removed from the Pulse signal dataset. Therefore, low-quality Pulse signal detection plays an important role in computational Pulse Diagnosis especially in the real-time Pulse monitoring. In this work, we focus on the detection of two common Pulse corruption types, i.e., saturation and artifact. For the detection of saturation, we use two criteria from its definition. For the artifact detection, we transform the Pulse signal into a complex network and detect the artifact by measuring the connectivity of the network. The experimental results show that the saturation and artifact detection method can both achieve better detection accuracy and better time resolution.
-
introduction computational Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:Pulse Diagnosis is a traditional Diagnosis technique by analyzing the tactile radial arterial palpation by trained fingertips; however it is a subjective skill which needs years of training and practice to master. Computational Pulse Diagnosis intends to employ some modern sensor and computer technology to make Pulse Diagnosis more objective. In this chapter, we will give an overview of computational Pulse Diagnosis. Firstly, the principle of Pulse Diagnosis and the traditional Pulse Diagnosis were introduced, and then the main concept of and the four stages of computational Pulse Diagnosis were introduced.
-
characterization of inter cycle variations for wrist Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:Although Pulse signal is quasiperiodic, most feature extraction methods usually consider it as a whole or only use a single cycle, neglecting the variations between Pulse cycles. To characterize both the inter- and intra-cycle variations, in this chapter we propose three feature extraction methods, i.e., simple combination, multi-scale entropy, and complex network. The simple combination method is a direct extension of conventional single-cycle feature extraction method by concatenating features from multiple cycles. The multi-scale entropy method measures the inter- and intra-cycle variations using entropies of different scales. The complex network method transforms the Pulse signal from time domain to network domain and measures the inter-cycle variations using the statistical properties on complex network. Experimental results show that the presented features are effective in characterizing both inter- and intra-cycle variations and can obtain better performance in Pulse Diagnosis.
-
edit distance for Pulse Diagnosis
2018Co-Authors: David Zhang, Wangmeng Zuo, Peng WangAbstract:In this chapter, by referring to the edit distance with real penalty (ERP) and the recent progress in k-nearest neighbors (KNN) classifiers, we propose two novel ERP-based KNN classifiers. Taking advantage of the metric property of ERP, we first develop an ERP-induced inner product and a Gaussian ERP kernel, then embed them into difference-weighted KNN classifiers, and finally develop two novel classifiers for Pulse waveform classification. The experimental results show that the proposed classifiers are effective for accurate classification of Pulse waveform.
-
Comparison of Three Different Types of Wrist Pulse Signals by Their Physical Meanings and Diagnosis Performance
IEEE Journal of Biomedical and Health Informatics, 2016Co-Authors: Peng Wang, David ZhangAbstract:Increasing interest has been focused on computational Pulse Diagnosis where sensors are developed to acquire Pulse signals, and machine learning techniques are exploited to analyze health conditions based on the acquired Pulse signals. By far, a number of sensors have been employed for Pulse signal acquisition, which can be grouped into three major categories, i.e., pressure, photoelectric, and ultrasonic sensors. To guide the sensor selection for computational Pulse Diagnosis, in this paper, we analyze the physical meanings and sensitivities of signals acquired by these three types of sensors. The dependence and complementarity of the different sensors are discussed from both the perspective of cardiovascular fluid dynamics and comparative experiments by evaluating disease classification performance. Experimental results indicate that each sensor is more appropriate for the Diagnosis of some specific disease that the changes of physiological factors can be effectively reflected by the sensor, e.g., ultrasonic sensor for diabetes and pressure sensor for arteriosclerosis, and improved Diagnosis performance can be obtained by combining three types of signals.
Chinghsing Luo - One of the best experts on this subject based on the ideXlab platform.
-
non invasive holistic health measurements using Pulse Diagnosis i validation by three dimensional Pulse mapping
European Journal of Integrative Medicine, 2016Co-Authors: Chinghsing Luo, Tingyi Huang, Chengying ChungAbstract:Abstract Introduction Pulse Diagnosis (PD) in Chinese medicine (CM) is a well-established clinical tool used to aid holistic Diagnosis. However, most of the recent efforts focusing on single-point Pulse waves (SPPWs) have been unable to successfully measure and translate how practitioners feel and interpret the Pulse. Methods A prerequisite to validating the holistic model used in CM requires Pulse-taking standards and a method to quantify the Pulse feeling during PD. Regarding the former, the flatness of the wrist radial artery and the Pulse-taking depth as defined by width of the radial artery were proposed. Regarding the latter, three-dimensional Pulse mapping (3DPM) was obtained using an array sensor to visibly and quantitatively measure the palpation of the Pulse. Cold stimulation was used to incite local vascular stiffness and the increase of the systematic blood pressure, to create the string-like Pulse feeling, in contrast to the normal gentle Pulse feeling. Results Three types of 3DPM were found, namely gentle, string-like, and slippery Pulse mappings. All SPPWs had similar one-dimensional shapes whereas the corresponding 3DPMs had quite different three-dimensional shapes. For gentle Pulse mapping (indicating health), an initial peak appeared and reached the top at the upstroke peak of SPPWs. As the local artery became stiffer by applying cold stimulation, the peak was replaced by a string-like shape. Additionally, a slippery 3DPM was recorded, with two peaks appearing sequentially like a bead flowing through the artery under a fingertip. Conclusion 3DPM provides quantitatively and non-invasively accesses information to inform holistic health assessment compared to SPPWs. CM can be improved and further developed quantifying Pulse palpation and verify its use in assessing a CM clinical holistic Diagnosis.
-
using an array sensor to determine differences in Pulse Diagnosis three positions and nine indicators
European Journal of Integrative Medicine, 2014Co-Authors: Yuwe Chu, Yufeng Chung, Chinghsing Luo, Chengchang YehAbstract:Abstract Introduction In Pulse Diagnosis, three positions ( Cun , Guan , Chi ) and three levels or depths ( Fu , Zhong , Chen ), called the Three Positions and Nine Indicators (TPNI) are generally used as a finger-reading method, to obtain a holistic view of the patient. However, single Pulse signals at nine TPNI locations (three depths at three positions) look quite similar in a waveform. Hence, the aim of this study was to determine if there was a significant difference between Pulse-taking depths (Fu, Zhong, and Chen) and Pulse-taking positions (Cun, Guan, and Chi). Method To explore the significance of array Pulses at the nine TPNI locations, a Bi-Sensing Pulse Diagnosis Instrument (BSPDI) with array sensors was used to measure wrist artery Pulse signals. It was proposed that a three-dimensional Pulse mapping (3DPM) could present array Pulses mimicking the fingertips’ sensations of a physician. Four parameters, namely peak value ( P _ V max ), power ( P _ P ), ascending slope ( P _AS), and descending slope ( P _DS) were elucidated from 3DPM using a two-way analysis of variance. Eight volunteers with TPNI health rule of thumb from the R.O.C. Air Force Academy participated in this research. Results The variance of four parameters at nine TPNI locations all reached the level of significance ( p Conclusions The differences in wrist artery signals exist between TPNI locations. TPNI Pulse Diagnosis could be used to check the holistic health of a patient as determined by TCM.
-
how to standardize the Pulse taking method of traditional chinese medicine Pulse Diagnosis
Computers in Biology and Medicine, 2013Co-Authors: Yufeng Chung, Chengchang Yeh, Chinghsing LuoAbstract:The aim of this report is to propose standard Pulse taking procedure of Traditional Chinese Medicine Pulse Diagnosis. In order to acquire full information from taking a wrist Pulse, this proposal adopts a tactile sensor with 12 sensing points at one sensing position, such as Cun, Guan, or Chi. Simultaneously Palpation (SP) and Pressing with One Finger (PWOF) are adopted to explore their differences of the detected Pulse signals. According to vertical dynamic characteristics, the results of a Pearson product moment reveal that the correlation coefficients of PWOF and SP are highly correlated from Fu to Chen. In addition, according to unique characteristics of body state, the results of a paired samples t test reveal that the SP and PWOF are indifferent at a specific Pulse taking depth. Hence, if using the Pulse-taking instrument with tactile sensors, it is concluded that Pulse signals taken by familiar SP and PWOF methods are shown in statistical indifferences among seven parameters (V"p"p"m"e"a"n",V"p"p"m"a"x, HR, LENGTH, WIDTH, AS, and DS).
-
possibility of quantifying tcm finger reading sensations i bi sensing Pulse Diagnosis instrument
European Journal of Integrative Medicine, 2012Co-Authors: Chinghsing Luo, Yufeng Chung, Chengchang Yeh, Da Hsua Feng, Yungchu Lee, Shi I Huang, Shu Ming Yeh, Chi Hsie LiangAbstract:Abstract Aim of this study This paper reports the construction and functionality of a newly designed Bi-Sensing Pulse Diagnosis Instrument (BSPDI) with a Pressure-Displacement Bi-Sensing System (PDBSS) coupled to a robot finger system. The BSPDI is used to simulate the three positions and nine indicators (TPNI) using the finger-reading rule, which addresses the three levels of superficial ( Fu ), medium ( Zhong ), and deep ( Chen ) at the three positions of distal ( CUN ), middle ( GUAN ) and proximal ( CHI ) on the wrist during Pulse Diagnosis. Materials and methods A strain gauge and polyvinylidene fluoride pressure sensor were integrated as a PDBSS to separately record finger-reading skills and sense a physician's fingertip sensations. The corresponding sensation and Pulse-taking displacement between the physician's fingertips and the BSPDI and the relationship between the Pulse Diagnosis and the nine TPNI displacements were used to determine the feasibility of the BSPDI. Results TPNI displacements, representing a physician's finger-reading skill, were recorded using a PDBSS. In this scenario, the BSPDI robot fingertips with pressure sensors were placed at the nine recorded TPNI displacements. The pressure sensors recorded Pulse signals to simulate a physician's fingertip sensations. Two groups of participants with different health statuses were evaluated by both the BSPDI and TCM physicians. Both groups received similar diagnoses, and the results of a two-way ANOVA indicated that Pulse diagnoses were highly related to the clinical experiences of the Pulse-taking displacements and positions. Conclusions The novel BSPDI can be used to successfully mimic a TCM physician's finger-reading skill, as this technique obtained Pulse signals at the nine TPNI displacements representing fingertip sensations.
-
exploring the conventional Pulse conditions using bi sensing Pulse Diagnosis instrument
BioMedical Engineering and Informatics, 2011Co-Authors: Yufeng Chung, Yuwe Chu, Chinghsing Luo, Chengchang YehAbstract:Pulse Diagnosis is one of the efficient techniques to detect the health status of a patient. However, Pulse Diagnosis suffers from a viewpoint of nonscientific and subjectivity. The goal of this report wants to propose an outline to integrate both the quantification of Pulse Diagnosis and application of clinic. Hence, the definitions of the quantifiable parameter are followed by clinical experiences. These core parameters are classified into four categories: position, rate and rhythm, shape, and trend. Using Bi-Sensing Pulse Diagnosis Instrument (BSPDI) to acquire wrist artery information for interpreting Pulse conditions. The DS (Depth-Strength) curve can be classified the Pulse condition whether it belongs to floating Pulse or sunken Pulse; and similarly, DR (Depth-Rate-Rhythm) curve to rapid Pulse or slow Pulse, DW (Depth-Width) curve to fine Pulse or large Pulse, DL (Depth-Length) curve to long Pulse or short Pulse. The Pulse conditions will be understood based on these curves and Pulse Diagnosis is no more subjectivity and nonscientific.
Guangming Lu - One of the best experts on this subject based on the ideXlab platform.
-
radial artery Pulse waveform analysis based on curve fitting using discrete fourier series
Computer Methods and Programs in Biomedicine, 2019Co-Authors: Zhixing Jiang, David Zhang, Guangming LuAbstract:Abstract Background and objectives: Radial artery Pulse Diagnosis has been playing an important role in traditional Chinese medicine (TCM). For its non-invasion and convenience, the Pulse Diagnosis has great significance in diseases analysis of modern medicine. The practitioners sense the Pulse waveforms in patients’ wrist to make diagnoses based on their non-objective personal experience. With the researches of Pulse acquisition platforms and computerized analysis methods, the objective study on Pulse Diagnosis can help the TCM to keep up with the development of modern medicine. Methods: In this paper, we propose a new method to extract feature from Pulse waveform based on discrete Fourier series (DFS). It regards the waveform as one kind of signal that consists of a series of sub-components represented by sine and cosine (SC) signals with different frequencies and amplitudes. After the Pulse signals are collected and preprocessed, we fit the average waveform for each sample using discrete Fourier series by least squares. The feature vector is comprised by the coefficients of discrete Fourier series function. Results: Compared with the fitting method using Gaussian mixture function, the fitting errors of proposed method are smaller, which indicate that our method can represent the original signal better. The classification performance of proposed feature is superior to the other features extracted from waveform, liking auto-regression model and Gaussian mixture model. Conclusions: The coefficients of optimized DFS function, who is used to fit the arterial pressure waveforms, can obtain better performance in modeling the waveforms and holds more potential information for distinguishing different psychological states.
