Deep learning models are widely used in the automatic analysis of radiological images. These techniques can train the weights of networks on large datasets as well as fine tuning the weights of pre...

Classification of the COVID-19 infected patients using DenseNet201 based deep transfer learning.

A survey on deep learning based face recognition

The COVID-19 infection is increasing at a rapid rate, with the availability of limited number of testing kits. Therefore, the development of COVID-19 testing kits is still an open area of research. Recently, many studies have shown that chest Computed Tomography (CT) images can be used for COVID-19 testing, as chest CT images show a bilateral change in COVID-19 infected patients. However, the classification of COVID-19 patients from chest CT images is not an easy task as predicting the bilateral change is defined as an ill-posed problem. Therefore, in this paper, a deep transfer learning technique is used to classify COVID-19 infected patients. Additionally, a top-2 smooth loss function with cost-sensitive attributes is also utilized to handle noisy and imbalanced COVID-19 dataset kind of problems. Experimental results reveal that the proposed deep transfer learning-based COVID-19 classification model provides efficient results as compared to the other supervised learning models.

Deep Transfer Learning Based Classification Model for COVID-19 Disease.

Image denoising review: From classical to state-of-the-art approaches

The COVID-19 infection is increasing at a rapid rate, with the availability of limited number of testing kits. Therefore, the development of COVID-19 testing kits is still an open area of research. Recently, many studies have shown that chest Computed Tomography (CT) images can be used for COVID-19 testing, as chest CT images show a bilateral change in COVID-19 infected patients. However, the classification of COVID-19 patients from chest CT images is not an easy task as predicting the bilateral change is defined as an ill-posed problem. Therefore, in this paper, a deep transfer learning technique is used to classify COVID-19 infected patients. Additionally, a top-2 smooth loss function with cost-sensitive attributes is also utilized to handle noisy and imbalanced COVID-19 dataset kind of problems. Experimental results reveal that the proposed deep transfer learning-based COVID-19 classification model provides efficient results as compared to the other supervised learning models. 

Deep Transfer Learning Based Classification Model for COVID-19 Disease

A drug-drug interaction or drug synergy is extensively utilised for cancer treatment. However, prediction of drug-drug interaction is defined as an ill-posed problem, because manual testing is only implementable on small group of drugs. Predicting the drug-drug interaction score has been a popular research topic recently. Recently many machine learning models have proposed in the literature to predict the drug-drug interaction score efficiently. However, these models suffer from the over-fitting issue. Therefore, these models are not so-effective for predicting the drug-drug interaction score. In this work, an integrated convolutional mixture density recurrent neural network is proposed and implemented. The proposed model integrates convolutional neural networks, recurrent neural networks and mixture density networks. Extensive comparative analysis reveals that the proposed model significantly outperforms the competitive models.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-syb.2019.0116

Efficient prediction of drug–drug interaction using deep learning models

A novel pose-invariant face recognition method is proposed by combining curvelet-invariant moments with curvelet neural network. First a special set of statistical coefficients using higher-order moments of curvelet are extracted as the feature vector and then the invariant features are fed into curvelet neural networks. Finally, supervised invariant face recognition is achieved by converging the neural network using curvelet as the activation function of the hidden layer neurons. The experimental results demonstrate that curvelet higher-order moments and curvelet neural networks achieve higher accuracy for face recognition across pose and converge rapidly than standard back propagation neural networks.

https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-bmt.2013.0019

Pose-invariant face recognition using curvelet neural network

Efficient face recognition using wavelet-based generalized neural network

Wireless sensor network consists of hundreds or thousands of low cost, low power, and self-organizing tiny sensor nodes that are deployed within the sensor network. Sensor network is susceptible to physical attacks due to deprived power and restricted resource capability and is exposed to external environment for transmitting and receiving data. Node capture attack is one of the most menacing attack in the wireless sensor network and may be physically captured by an adversary for extracting confidential information regarding cryptographic keys, node’s unique id, and so forth, from its memory to eliminate the confidentiality and integrity of the wireless links. Node capture attack suffers from severe security breach and tremendous network cost. We propose an empirically designed multiple objectives node capture attack algorithm based on optimization functions as an effective solution against the attacking efficiency of node capture attack. Finding robust assailant optimization-particle swarm optimization and genetic algorithm (FiRAO-PG) consists of multiple objectives: maximum node participation, maximum key participation, and minimum resource expenditure to find optimal nodes using PSO and GA. It will leverage a comprehensive tool to destroy maximum portion of the network realizing cost-effectiveness and higher attacking efficiency. The simulation results manifest that FiRAO-PG can provide higher fraction of compromised traffic than matrix algorithm (MA) so the attacking efficiency of FiRAO-PG is higher.

/pdf/finding-robust-assailant-using-optimization-functions-firao-45r7fsk9bc.pdf

Finding Robust Assailant Using Optimization Functions (FiRAO-PG) in Wireless Sensor Network

Image denoising using wavelet transform has been successful as wavelet transform generates a large number of small coefficients and a small number of large coefficients. Basic denoising algorithm that using the wavelet transform consists of three steps - first computing the wavelet transform of the noisy image, thresholding is performed on the detail coefficients in order to remove noise and finally inverse wavelet transform of the modified coefficients is taken. This paper reviews the state of art methods of image denoising using wavelet thresholding. An Experimental analysis of wavelet based methods Visu Shrink, Sure Shrink, Bayes Shrink, Prob Shrink, Block Shrink and Neigh Shrink Sure is performed. These wavelet based methods are also compared with spatial domain methods like median filter and wiener filter. Results are evaluated on the basis of Peak Signal to Noise Ratio and visual quality of images. In the experiment, wavelet based methods perform better than spatial domain methods. In wavelet domain, recent methods like prob shrink, block shrink and neigh shrink sure performed better as compared to other wavelet based methods.

/pdf/a-comparative-study-of-wavelet-thresholding-for-image-2hyeyq87u0.pdf

Poonam Sharma

Papers

Efficient prediction of drug–drug interaction using deep learning models

Pose-invariant face recognition using curvelet neural network

Efficient face recognition using wavelet-based generalized neural network

Finding Robust Assailant Using Optimization Functions (FiRAO-PG) in Wireless Sensor Network

A Comparative Study of Wavelet Thresholding for Image Denoising