Adaptive k-means clustering algorithm for MR breast image segmentation

Image segmentation is vital for meaningful analysis and interpretation of the medical images. The most popular method for clustering is k-means clustering. This article presents a new approach intended to provide more reliable magnetic resonance (MR) breast image segmentation that is based on adaptation to identify target objects through an optimization methodology that maintains the optimum result during iterations. The proposed approach improves and enhances the effectiveness and efficiency of the traditional k-means clustering algorithm. The performance of the presented approach was evaluated using various tests and different MR breast images. The experimental results demonstrate that the overall accuracy provided by the proposed adaptive k-means approach is superior to the standard k-means clustering technique.     

A computational knowledge representation model for cognitive computers

The accumulating data are easy to store but the ability of understanding and using it does not keep track with its growth. So researches focus on the nature of knowledge processing in the mind. This paper proposes a semantic model (CKRMCC) based on cognitive aspects that enables cognitive computer to process the knowledge as the human mind and find a suitable representation of that knowledge. In cognitive computer, knowledge processing passes through three major stages: knowledge acquisition and encoding, knowledge representation, and knowledge inference and validation. The core of CKRMCC is knowledge representation, which in turn proceeds through four phases: prototype formation phase, discrimination phase, generalization phase, and algorithm development phase. Each of those phases is mathematically formulated using the notions of real-time process algebra. The performance efficiency of CKRMCC is evaluated using some datasets from the well-known UCI repository of machine learning datasets. The acquired datasets are divided into training and testing data that are encoded using concept matrix. Consequently, in the knowledge representation stage, a set of symbolic rule is derived to establish a suitable representation for the training datasets. This representation will be available in a usable form when it is needed in the future. The inference stage uses the rule set to obtain the classes of the encoded testing datasets. Finally, knowledge validation phase is validating and verifying the results of applying the rule set on testing datasets. The performances are compared with classification and regression tree and support vector machine and prove that CKRMCC has an efficient performance in representing the knowledge using symbolic rules.     

Nonlinear features of heart rate variability in paranoid schizophrenic

Cardiovascular mortality is significantly increased in patients suffering from schizophrenia. However, psychotic symptoms are quantified by means of the scale for the assessment of positive and negative symptoms, but many investigations try to introduce new etiology for psychiatric disorders based on combination of biological, psychological and social causes. Classification between healthy and paranoid cases has been achieved by time, frequency, Hilbert–Huang (HH) and a combination between those features as a hybrid features. Those features extracted from the Hilbert–Huang transform for each intrinsic mode function (IMF) of the detrended time series for each healthy case and paranoid case. Short-term ECG recordings of 20 unmedicated patients suffering from acute paranoid schizophrenia and those obtained from healthy matched peers have been utilized in this investigation. Frequency features: very low frequency (VLF), low frequency (LF), high frequency (HF) and HF/LF (ratio) produced promising success rate equal to 97.82 % in training and 97.77 % success rate in validation by means of IMF1 and ninefolds. Time–frequency features [LF, HF and ratio, mean, maximum (max), minimum (min) and standard deviation (SD)] provided 100 % success in both training and validation trials by means of ninefolds for IMF1 and IMF2. By utilizing IMF1 and ninefolds, frequency and Hilbert–Hang features [LF, HF, ratio, mean value of envelope ( \(\bar{a}\) )] produced 96.87 and 95.5 % for training and validation, respectively. By analyzing the first IMF and using ninefolds, time and Hilbert–Hang features [mean, max, min, SD, median, first quartile (Q1), third quartile (Q3), kurtosis, skewness, Shannon entropy, approximate entropy and energy, ( \(\bar{a}\) ), level of envelope variation ([ \(\dot{a}\) (t)]^2), central frequency \((\bar{W})\) and number of zero signal crossing \((\left| {\bar{W}} \right|)\) ] produced a 100 % success rate in training and 90 % success rate in validation. Time, frequency and HH features [energy, VLF, LF, HF, ratio and ( \(\bar{a}\) )] provided 97.5 % success rate in training and 95.24 % success rate in validation using IMF1 and sixfolds. However, frequency features have produced promising classification success rate, but hybrid features emerged the highest classification success rate than using features in each domain separately.     

Supervised hybrid feature selection based on PSO and rough sets for medical diagnosis

Medical datasets are often classified by a large number of disease measurements and a relatively small number of patient records. All these measurements (features) are not important or irrelevant/noisy. These features may be especially harmful in the case of relatively small training sets, where this irrelevancy and redundancy is harder to evaluate. On the other hand, this extreme number of features carries the problem of memory usage in order to represent the dataset. Feature Selection (FS) is a solution that involves finding a subset of prominent features to improve predictive accuracy and to remove the redundant features. Thus, the learning model receives a concise structure without forfeiting the predictive accuracy built by using only the selected prominent features. Therefore, nowadays, FS is an essential part of knowledge discovery. In this study, new supervised feature selection methods based on hybridization of Particle Swarm Optimization (PSO), PSO based Relative Reduct (PSO-RR) and PSO based Quick Reduct (PSO-QR) are presented for the diseases diagnosis. The experimental result on several standard medical datasets proves the efficiency of the proposed technique as well as enhancements over the existing feature selection techniques.     

Microstructure of KTaxNb1−xO3 Thin Films on MgO (100) Single Crystals

Thin KTa_xNb_1−xO₃ (KTN) films were prepared by deposition of sol–gel precursor solutions on MgO (100) single crystals. Crystal structure and microstructure of the films as a function of processing parameters, such as rate, duration, and temperature of postdeposition heat treatment, were studied. Several techniques such as X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) were employed to analyze the films. It was observed that slow heating of KTN films promotes pyrochlore formation while fast-firing of the films results in predominant formation of the perovskioe phase. In slow-heated samples, TEM showed randomly oriented pyrochlore crystallites with a vermicular nanostructure of 10–30 nm with an interpenetrating porosity of the same range. In fast-fired samples, large perovskite pockets with pyrochlore crystallites scattered among them were seen. The large perovskite grains were on the order of 0.1–0.5 μm, irregular in shape and porous. Transmission electron diffraction indicated these were single crystals, and ferroelectric domains were observed in them. Films of up to 1 μm thick were obtained by multiple deposition of the sol–gel KTN. Dense films were achieved when each layer was densified at 750°C for 2 h before the next layer was deposited.     

Motion-compensated data decomposition algorithm to accelerate dynamic cardiac MRI

Objectives

In dynamic cardiac magnetic resonance imaging (MRI), the spatiotemporal resolution is often limited by low imaging speed. Compressed sensing (CS) theory can be applied to improve imaging speed and spatiotemporal resolution. The combination of compressed sensing and low-rank matrix completion represents an attractive means to further increase imaging speed. By extending prior work, a Motion-Compensated Data Decomposition (MCDD) algorithm is proposed to improve the performance of CS for accelerated dynamic cardiac MRI.

Materials and methods

The process of MCDD can be described as follows: first, we decompose the dynamic images into a low-rank (L) and a sparse component (S). The L component includes periodic motion in the background, since it is highly correlated among frames, and the S component corresponds to respiratory motion. A motion-estimation/motion-compensation (ME-MC) algorithm is then applied to the low-rank component to reconstruct a cardiac motion compensated dynamic cardiac MRI.

Results

With validations on the numerical phantom and in vivo cardiac MRI data, we demonstrate the utility of the proposed scheme in significantly improving compressed sensing reconstructions by minimizing motion artifacts. The proposed method achieves higher PSNR and lower MSE and HFEN for medium to high acceleration factors.

Conclusion

The proposed method is observed to yield reconstructions with minimal spatiotemporal blurring and motion artifacts in comparison to the existing state-of-the-art methods.

     

Comparison of different fluid dynamics in activated sludge system for the treatment of a stimulated milk processing wastewater: Process analysis and optimization

Wastewater from the milk industry usually undergoes activated sludge ahead of refining treatments, final discharge or reuse. To identify the most effective bioreactor hydraulic regime for the secondary treatment of wastewater resulting from the milk industry in an activated sludge system, two lab-scale activated sludge systems characterized by a different configuration and fluid dynamics (i.e., a compartmentalized activated sludge (CAS) with plug flow regime and a complete mixed activated sludge (AS)) were operated in parallel, inoculated with the same microbial consortium and fed with identical streams of a stimulated dairy wastewater. The effect of three process and operational variables??influent chemical oxygen demand (COD) concentration, sludge recycle ratio (R) and hydraulic retention time (HRT)??on the performance of the two systems were investigated. Experiments were conducted based on a central composite face-centered design (CCFD) and analyzed using response surface methodology (RSM). The region of exploration for treatment of the synthetic wastewater was taken as the area enclosed by the COD _in (200, 1,000 mg/l), R (1, 5), and HRT (2, 5 h) boundaries. To evaluate the process, three parameters, COD removal efficiency (E), specific substrate utilization rate (U), and sludge volume index (SVI), were measured and calculated over the course of the experiments as the process responses. The change of the flow regime from complete-mix to plug flow resulted in considerable improvements in the COD removal efficiency of milk wastewater and sludge settling properties. SVI levels for CAS system (30?C58 ml/g) were considerably smaller that for the AS system (50?C145 ml/g). In addition, the biomass production yield could be reduced by about 10% compared to the AS system. The results indicated that for the wastewater, the design HRT of a CAS reactor could be shortened to 2?C4 h.