Early detection of disease has become a crucial problem due to rapid population growth in medical research in recent times. With the rapid population growth, the risk of death incurred by breast cancer is rising exponentially. Breast cancer is the second most severe cancer among all of the cancers already unveiled. An automatic disease detection system aids medical staffs in disease diagnosis and offers reliable, effective, and rapid response as well as decreases the risk of death. In this paper, we compare five supervised machine learning techniques named support vector machine (SVM), K-nearest neighbors, random forests, artificial neural networks (ANNs) and logistic regression. The Wisconsin Breast Cancer dataset is obtained from a prominent machine learning database named UCI machine learning database. The performance of the study is measured with respect to accuracy, sensitivity, specificity, precision, negative predictive value, false-negative rate, false-positive rate, F1 score, and Matthews Correlation Coefficient. Additionally, these techniques were appraised on precision-recall area under curve and receiver operating characteristic curve. The results reveal that the ANNs obtained the highest accuracy, precision, and F1 score of 98.57%, 97.82%, and 0.9890, respectively, whereas 97.14%, 95.65%, and 0.9777 accuracy, precision, and F1 score are obtained by SVM, respectively.

The eight-chain model, also known as ArrudaBoyce model, is widely used to capture the rate-independent hyperelastic response of rubber-like materials. The parameters of this model are physically based and explained from micromechanics of chain molecules. Despite its excellent performance with only two material parameters to capture bench measurements in uniaxial and pure shear regime, the model is known to be significantly deficient in predicting the equibiaxial data. To ameliorate such drawback, over the years, several modified versions of this successful model have been proposed in the literature. The so-called full-network model is another micromechanically motivated chain model, which has also few modified versions in the literature. For this study, two modified versions of the full-network model have been selected. In this contribution, five modified versions of the Arruda-Boyce model and two modified versions of full-network model are critically compared with the classical eight-chain model for their adequacy in representing equibiaxial data. To do a comparison of all selected models in reproducing the well-known Treloar data, the analytical expressions for the three homogeneous deformation modes, that is, uniaxial tension, equibiaxial tension, and pure shear have been derived and the performances of the selected models are analysed. The comparative study demonstrates that modified Flory-Erman model, Gornet-Desmorat (GD) model, Meissner-Matějka model, and bootstrapped eight-chain model predict well the three deformation modes compare to the classical eight-chain model.

No abstract

Watermarking technique is a method to protect ownership of digital multimedia. Most existing watermarking techniques achieve a good level of imperceptibility and robustness. The challenges to achieve higher invisibility and resistance with lower computational time motivate researchers to work on new watermarking schemes. Robustness against noise attacks and JPEG2000 compression needs to be improved to acquire a better resistance capability of the watermark. In this paper, we present a block-based Tchebichef watermarking technique for protecting copyrights. In this technique, the host image is first divided into non-overlapping blocks and Tchebichef moments are calculated for each block. The watermarks are embedded into the blocks with lower visual entropies. The watermark image is scrambled by Arnold transform before embedding into the Tchebichef moments of the selected image blocks. The proposed watermarking scheme was tested under noise additions, filtering, cropping and compressing attacks. Our scheme was verified and compared to the existing watermarking techniques under image geometric and processing attacks. Furthermore, the proposed scheme demonstrated a superior performance in robustness under noise attacks and JPEG2000. INDEX TERMS Watermarking, copyright protection, embedding algorithm, tchebichef moments, visual entropies.

No abstract