Mohamad Forouzanfar

Parameter optimization of improved fuzzy c-means clustering algorithm for brain MR image segmentation

A traditional approach to segmentation of magnetic resonance (MR) images is the fuzzy c-means (FCM) clustering algorithm. The efficacy of FCM algorithm considerably reduces in the case of noisy data. In order to improve the performance of FCM algorithm, researchers have introduced a neighborhood attraction, which is dependent on the relative location and features of neighboring pixels. However, determination of degree of attraction is a challenging task which can considerably affect the segmentation results. This paper presents a study investigating the potential of genetic algorithms (GAs) and particle swarm optimization (PSO) to determine the optimum value of degree of attraction. The GAs are best at reaching a near optimal solution but have trouble finding an exact solution, while PSOs-group interactions enhances the search for an optimal solution. Therefore, significant improvements are expected using a hybrid method combining the strengths of PSO with GAs, simultaneously. In this context, a hybrid GAs/PSO (breeding swarms) method is employed for determination of optimum degree of attraction. The quantitative and qualitative comparisons performed on simulated and real brain MR images with different noise levels demonstrate unprecedented improvements in segmentation results compared to other FCM-based methods.

Coefficient-Free Blood Pressure Estimation Based on Pulse Transit Time–Cuff Pressure Dependence

Oscillometry is a popular technique for automatic estimation of blood pressure (BP). However, most of the oscillometric algorithms rely on empirical coefficients for systolic and diastolic pressure evaluation that may differ in various patient populations, rendering the technique unreliable. A promising complementary technique for automatic estimation of BP, based on the dependence of pulse transit time (PTT) on cuff pressure (CP) (PTT-CP mapping), has been proposed in the literature. However, a theoretical grounding for this technique and a nonparametric BP estimation approach are still missing. In this paper, we propose a novel coefficient-free BP estimation method based on PTT-CP dependence. PTT is mathematically modeled as a function of arterial lumen area under the cuff. It is then analytically shown that PTT-CP mappings computed from various points on the arterial pulses can be used to directly estimate systolic, diastolic, and mean arterial pressure without empirical coefficients. Analytical results are cross-validated with a pilot investigation on ten healthy subjects where 150 simultaneous electrocardiogram and oscillometric BP recordings are analyzed. The results are encouraging whereby the mean absolute errors of the proposed method in estimating systolic and diastolic pressures are 5.31 and 4.51 mmHg, respectively, relative to the Food and Drug Administration approved Omron monitor. Our work thus shows promise toward providing robust and objective BP estimation in a variety of patients and monitoring situations.

Feature-Based Neural Network Approach for Oscillometric Blood Pressure Estimation

In this paper, we present a novel feature-based neural network (NN) approach for estimation of blood pressure (BP) from wrist oscillometric measurements. Unlike previous methods that use the raw oscillometric waveform envelope (OMWE) as input to the NN, in this paper, we propose to use features extracted from the envelope. The OMWE is mathematically modeled as a sum of two Gaussian functions. The optimum parameters of this model are found by minimizing the least squares error between the model and the OMWE using the Levenberg-Marquardt algorithm and are used as features. Two separate feed-forward NNs (FFNNs) are then designed to estimate the systolic and diastolic BPs using these features. The FFNNs are trained using the resilient backpropagation learning algorithm and tested on a data set of BP measurements recorded from 85 subjects. The performance is then compared with that of the conventional maximum amplitude algorithm, adaptive neuro-fuzzy inference system, and already published NN-based methods. It is found that the proposed approach achieves lower values of mean absolute error and standard deviation of error in the estimation of BP. In addition, the proposed approach has the following advantages: lower complexity with respect to the design parameters, smaller training data set, and lower computational load.

Oscillometric Blood Pressure Estimation: Past, Present, and Future

The use of automated blood pressure (BP) monitoring is growing as it does not require much expertise and can be performed by patients several times a day at home. Oscillometry is one of the most common measurement methods used in automated BP monitors. A review of the literature shows that a large variety of oscillometric algorithms have been developed for accurate estimation of BP but these algorithms are scattered in many different publications or patents. Moreover, considering that oscillometric devices dominate the home BP monitoring market, little effort has been made to survey the underlying algorithms that are used to estimate BP. In this review, a comprehensive survey of the existing oscillometric BP estimation algorithms is presented. The survey covers a broad spectrum of algorithms including the conventional maximum amplitude and derivative oscillometry as well as the recently proposed learning algorithms, model-based algorithms, and algorithms that are based on analysis of pulse morphology and pulse transit time. The aim is to classify the diverse underlying algorithms, describe each algorithm briefly, and discuss their advantages and disadvantages. This paper will also review the artifact removal techniques in oscillometry and the current standards for the automated BP monitors.

Block-wise 2D kernel PCA/LDA for face recognition

Direct extension of (2D) matrix-based linear subspace algorithms to kernel-induced feature space is computationally intractable and also fails to exploit local characteristics of input data. In this letter, we develop a 2D generalized framework which integrates the concept of kernel machines with 2D principal component analysis (PCA) and 2D linear discriminant analysis (LDA). In order to remedy the mentioned drawbacks, we propose a block-wise approach based on the assumption that data is multi-modally distributed in so-called block manifolds. Proposed methods, namely block-wise 2D kernel PCA (B2D-KPCA) and block-wise 2D generalized discriminant analysis (B2D-GDA), attempt to find local nonlinear subspace projections in each block manifold or alternatively search for linear subspace projections in kernel space associated with each blockset. Experimental results on ORL face database attests to the reliability of the proposed block-wise approach compared with related published methods.

MRI fuzzy segmentation of brain tissue using IFCM algorithm with particle swarm optimization

Medical image segmentation is a complex and challenging task due to the intrinsic nature of the images. Magnetic resonance imaging (MRI) segmentation is of particular importance for further image analysis. Fuzzy c-mean (FCM) is a common clustering algorithm which is used for segmentation of MR images. However in the case of noisy MR images, efficiency of this algorithm considerably reduces. Recently, researchers have introduced two new parameters in order to improve the performance of traditional FCM in the case of noisy images. New parameters are computed using artificial neural networks and through a complex and time consuming optimization problem. In this paper, we present a new method for computation of these two parameters, efficiently. We use a particle swarm optimization (PSO) method and show the capability of PSO to find optimal values of these parameters. The advantage of the new proposed method is its simplified computations. Our simulation results on a set of noisy MR images, demonstrate the effectiveness of proposed optimization method compared with some related recent algorithms.

Mathematical modeling and parameter estimation of blood pressure oscillometric waveform

In this paper, a mathematical model for the blood pressure oscillometric waveform (OMW) is developed and a statespace approach using the extended Kalman filter (EKF) is proposed to adaptively estimate and track parameters of clinical interest. The OMW model is driven by a previously proposed pressure-lumen area model of the artery under the deflating cuff. The arterial lumen area is a function of vessel properties, the cuff pressure, and the arterial pressure. Over the deflation period, the arterial pressure causes lumen area oscillations while the deflating cuff pressure adds a slow-varying component to these oscillations. In the previous literature, it has been demonstrated that the oscillometric pulses are proportional to the arterial area oscillations. In this paper, the OMW is modeled as the difference between the whole lumen area model and the slow-varying component of the lumen area caused by the deflating cuff pressure. The OMW model is then represented in the statespace and the extended Kalman filter (EKF) is incorporated to estimate and track the time-varying model parameters during the cuff deflation period. The parameter tracking performance of the EKF is evaluated on a simulated noisy OMW.

Oscillometric blood pressure estimation using principal component analysis and neural networks

Estimation of systolic and diastolic pressures from the oscillometric waveform is a challenging task in noninvasive electronic blood pressure (BP) monitoring devices. Since the conventional oscillometric algorithms cannot model and extract the complex and nonlinear relationship that may exist between BP and oscillometric waveform, artificial neural networks (NNs) have been proposed as a possible alternative. However, the research on this topic has been limited to some simple architectures that directly estimate the BP from raw oscillation amplitudes (OAs). In this paper, we propose principal component analysis as a preprocessing step to decorrelate the OAs and extract the most effective components. Two architectures of NNs, namely, feed-forward and cascade-forward, are employed to estimate the BP using the preprocessed OAs. The networks are trained using the gradient descent with momentum and adaptive learning rate backpropagation algorithm and tested on a dataset of 85 BP waveforms. The performance is then compared with that of the conventional maximum amplitude algorithm and already published NN-based methods. It is found that the proposed networks achieve lower values of mean absolute error and standard deviation of error in estimation of BP compared with the other studied methods.

Adaptive neuro-fuzzy inference system for oscillometric blood pressure estimation

This paper presents a novel approach using principal component analysis (PCA) and adaptive neuro-fuzzy inference system (ANFIS) for estimation of blood pressure (BP) from oscillometric waveforms. The proposed method consists of three stages. In the first stage, the oscillation amplitudes (OAs) of the oscillometric waveforms are represented as a function of the cuff pressure. In the second stage, the PCA is utilized to reduce the dimensionality of the input space by extracting the most effective features from the OAs. Finally, in the third stage, the ANFIS is employed to perform the BP estimation. The proposed method is tested on a dataset collected from 85 subjects and the results are compared with conventional maximum amplitude algorithm and published neural network-based methods. It is found that the proposed method achieves lower values of mean absolute error and standard deviation of error in estimation of BP compared with the other studied methods.

Ratio-Independent Blood Pressure Estimation by Modeling the Oscillometric Waveform Envelope

Oscillometry is the most common measurement method used in electronic blood pressure (BP) monitors. However, most of the existing oscillometric algorithms employ empirical ratios on the oscillometric waveform envelope (OMWE) to determine the systolic pressure (SP) and diastolic pressure (DP). As these algorithms do not consider the cardiovascular system parameters that may vary due to health conditions or age, the pressure estimates are not always reliable. In this paper, we develop a new mathematical model for the OMWE by incorporating an existing model of the cuff-arm-artery system. The unique feature of our developed model is that it explicitly represents the relationship between the OMWE and the SP and DP. Based on our developed model, we propose a new ratio-independent oscillometric BP estimation method. The proposed method is based on minimizing the sum of the squared errors between our model and the OMWE using the trust-region-reflective algorithm. Our proposed method is validated in a pilot study against Omron HEM-790IT and BpTRU BMP-100 BP monitors. It is found that the mean absolute error of the proposed method in estimating SP and DP is 4.60 and 4.53 mmHg, respectively, relative to the Omron monitor, and 3.66 and 2.84 mmHg, respectively, relative to the BpTRU monitor. The proposed model thus shows promise toward developing robust BP estimation methods.