A pot experiment was carried out to evaluate the effect of clay minerals, compost and the interaction between them (Zeolite = Z, Bentonite = B, compost = C, zeolite + bentonite = ZB, zeolite + compost = ZC and bentonite + compost = BC) as soil amendments and their effect on reducing heavy metals (HMs = Cd, Ni and Pb) accumulation in barley plant (Hordeum vulgare L. Var. Giza 132) and their mobility in two contaminated soils (Abou-Rawash, Giza Governorate and Kafr El-Sheikh Governorate). The previous soil amendments were add by three rates (0, 1and 2%). Barley plants were harvested at 90 day. All treatments led to increasing fresh and dry weight of barley plants as well as decrease its concentration and content of HMs compared with control. Zeolite produced the highest growth while the lowest accumulation of HMs was resulted by bentonite addition. Generally, clay minerals are able to improving growth of plants in polluted areas in addition to decreasing HMs uptake by plants in absence or presence of compost.

It has been proven that due to generation and recombination of charge carriers at the presence of an external constant electric field concentration fluctuations of charge carriers and electric field occur in semiconductors with deep traps. For the first time a Van der Pol type equation was obtained for these semiconductors for the alternating electric field. From the solution of the obtained equation both the amplitude and oscillation frequency were determined in the first approximation by the method of N.N. Bogoliubov and Y.A. Mitropolsky. It was shown that the frequency of oscillations in the first approxi-mation is more important than in the zero approximation. The amplitude of oscillations tends to a finite value at very high time value . This proves that there is a steady dynamic mode. A graph indicating the dependence of amplitude on time was developed. The values of the electric field and the constants of generation and recombination of charge carriers were determined.

In India, every year over 2 lakh people lose their lives due to liver disease. To diagnose the various types of liver diseases, the medical imaging based schemes have been used. In this paper, the ultrasound image has been considered due to their benefits like safe and low cost compared to overall medical imaging modalities for liver disease classification by using machine learning with optimization schemes. In this work, the Spiking Neuron based Neural Network (SNNN) with Cat Swarm Optimization (CSO) is proposed for classifying normal, fatty and heterogeneous liver images. The ultrasound images have been affected by the speckle noise, which is reduced by using bilateral filter in pre-processing. In order to increase detection accuracy, the Gray Level Co-Occurrence Matrix (GLCM), Linear Discriminant Analysis (LDA), Gray Level Different Statistics (GLDS) and statistical features are extracted. After that, the extracted features are combined and then the efficient features are selected by Dragon Fly (DF) algorithm for classification. Finally, the SNNN is used for liver disease classification. To reduce the pattern recognition issue and achieve better accuracy with low computational complexity, the proposed scheme introduced a bio inspired algorithm of CSO as a learning approach to training a SNNN. The simulation results demonstrate that the proposed SNNN-CSO has obtained better performance and less false error rate compared to exist the Support Vector Machine (SVM).