Automated system for characterization and classification of malaria‐infected stages using light microscopic images of thin blood smears

In this paper, we propose a comprehensive image characterization cum classification framework for malaria‐infected stage detection using microscopic images of thin blood smears. The methodology mainly includes microscopic imaging of Leishman stained blood slides, noise reduction and illumination correction, erythrocyte segmentation, feature selection followed by machine classification. Amongst three‐image segmentation algorithms (namely, rule‐based, Chan–Vese‐based and marker‐controlled watershed methods), marker‐controlled watershed technique provides better boundary detection of erythrocytes specially in overlapping situations. Microscopic features at intensity, texture and morphology levels are extracted to discriminate infected and noninfected erythrocytes. In order to achieve subgroup of potential features, feature selection techniques, namely, F‐statistic and information gain criteria are considered here for ranking. Finally, five different classifiers, namely, Naive Bayes, multilayer perceptron neural network, logistic regression, classification and regression tree (CART), RBF neural network have been trained and tested by 888 erythrocytes (infected and noninfected) for each features’ subset. Performance evaluation of the proposed methodology shows that multilayer perceptron network provides higher accuracy for malaria‐infected erythrocytes recognition and infected stage classification. Results show that top 90 features ranked by F‐statistic (specificity: 98.64%, sensitivity: 100%, PPV: 99.73% and overall accuracy: 96.84%) and top 60 features ranked by information gain provides better results (specificity: 97.29%, sensitivity: 100%, PPV: 99.46% and overall accuracy: 96.73%) for malaria‐infected stage classification.     

Improved cuckoo search with particle swarm optimization for classification of compressed images

The need for a general purpose Content Based Image Retrieval (CBIR) system for huge image databases has attracted information-technology researchers and institutions for CBIR techniques development. These techniques include image feature extraction, segmentation, feature mapping, representation, semantics, indexing and storage, image similarity-distance measurement and retrieval making CBIR system development a challenge. Since medical images are large in size running to megabits of data they are compressed to reduce their size for storage and transmission. This paper investigates medical image retrieval problem for compressed images. An improved image classification algorithm for CBIR is proposed. In the proposed method, RAW images are compressed using Haar wavelet. Features are extracted using Gabor filter and Sobel edge detector. The extracted features are classified using Partial Recurrent Neural Network (PRNN). Since training parameters in Neural Network are NP hard, a hybrid Particle Swarm Optimization (PSO) – Cuckoo Search algorithm (CS) is proposed to optimize the learning rate of the neural network.     

Generalized probability model for estimating the proportion of the female population at different levels of fertility based on census sampling from a mixed population

The present exercise is an attempt to generalize a probability model referring to the waiting time distribution for conception in a mixed group of fecund females with two varying levels of fecundity, based on a census sample truncated at time T =t from the date of resumption of fecundity following an earlier conception (Biswas, 1975, Sankhya, Series B, 37, Part III). The generalization is made in two aspects, viz. instead of fixed conception rates, time-dependent rates decreasing with the age of the woman have been used. Additionally, unlike the model of Biswas (1975) consisting of two different fecund groups, three different levels of fecundity have been taken into consideration in this paper. Estimates of the proportion of fecund females over time, viz. π_i(t) (i = 1, 2, 3) are attempted by the method of maximum likelihood on the basis of a census sample truncated at time T = t under the generalized set up.     

GEOMETRIC PROGRAMMING. A NEW HYBRID ALGORITHM BASED ON LIAPUNOV CONCEPTS (MADLAP)

This article concerns the development of a simple yet very efficient heuristic procedure for optimizing the design or a reactor system employing continuous stirred tanks in series using Michaelis-Menten kinetics. The heuristic procedure employs a modified form or the well known Golden Search method to estimate the optimal number of reactors to be employed. The procedure was tested on an IBM personal computer under a wide range of values for three model parameters, namely the economy or scale factor, the scaled Michaelis-Menten constant and the desired degree of conversion to be attained. The results of this study indicated that the heuristic optimization procedure is capable of obtaining solutions that are either identical or very close to the actual optimal solutions in a very efficient manner.     

Separately balanced realization of two-dimensional separable-denominator transfer functions†

In this paper a method is developed for the computation of a separately balanced state-space model for a given two-dimensional transfer function with separable denominator. The procedure relies on the eigenvalue/eigenvector decomposition (or singular-value decomposition) of two real symmetric matrices determined directly from the given transfer function.     

Cryogenic forming behaviour of AW-6016-T4 sheet

The objective of this work is to study the cryogenic sheet metal forming behaviour of precipitation hardening AW-6016-T4. In this regard, the flow curves and forming limit curves were obtained by tension and Nakazima experimental testing methods in the temperature ranges from ?196 to 25 °C. It was found that strength and elongation increase with decreasing temperature. Small but perceived differences between microstructure of the material deformed at the room and cryogenic temperatures respectively were identified by electron backscatter diffraction (EBSD) analysis. However, no significant difference in the precipitation kinetics during continuous heating in the DSC has been observed. This study has demonstrated the potential of cryogenic forming by manufacturing a B-pillar part with 8 mm depth of side design element as compared to 6 mm at room temperature.     

Experimental determination of mechanical parameters in sensorless vector-controlled induction motor drive

High-performance industrial drives widely employ induction motors with position sensorless vector control (SLVC). The state-of-the-art SLVC is first reviewed in this paper. An improved design procedure for current and flux controllers is proposed for SLVC drives when the inverter delay is significant. The speed controller design in such a drive is highly sensitive to the mechanical parameters of the induction motor. These mechanical parameters change with the load coupled. This paper proposes a method to experimentally determine the moment of inertia and mechanical time constant of the induction motor drive along with the load driven. The proposed method is based on acceleration and deceleration of the motor under constant torque, which is achieved using a sensorless vector-controlled drive itself. Experimental results from a 5-hp induction motor drive are presented.