Deep Learning Application for Effective Classification of Different Types of Psoriasis

Psoriasis is a chronic inflammatory skin disorder mediated by the immune response that affects a large number of people. According to latest worldwide statistics, 125 million individuals are suffering from psoriasis. Deep learning techniques have demonstrated success in the prediction of skin diseases and can also lead to the classification of different types of psoriasis. Hence, we propose a deep learning-based application for effective classification of five types of psoriasis namely, plaque, guttate, inverse, pustular, and erythrodermic as well as the prediction of normal skin. We used 172 images of normal skin from the BFL NTU dataset and 301 images of psoriasis from the Dermnet dataset. The input sample images underwent image preprocessing including data augmentation, enhancement, and segmentation which was followed by color, texture, and shape feature extraction. Two deep learning algorithms of convolutional neural network (CNN) and long short-term memory (LSTM) were applied with the classification models being trained with 80% of the images. The reported accuracies of CNN and LSTM are 84.2% and 72.3%, respectively. A paired sample T-test exhibited significant differences between the accuracies generated by the two deep learning algorithms with a p < 0.001 . The accuracies reported from this study demonstrate potential of this deep learning application to be applied to other areas of dermatology for better prediction.

Skin Lesion Segmentation in Dermoscopy Imagery

The main purpose of this study is to find an optimum method for segmentation of skin lesion images. In the present world, Skin cancer has proved to be the most deadly disease. The present research paper has developed a model which encompasses two gradations, the first being pre-processing for the reduction of unwanted artefacts like hair, illumination or many other by enhanced technique using threshold and morphological operations to attain higher accuracy and the second being segmentation by using k-mean with optimized Firefly Algorithm (FFA) technique. The online image database from the International Skin Imaging Collaboration (ISIC) archive dataset and dermatology service of Hospital Pedro Hispano (PH2) dataset has been used for input sample images. The parameters on which the proposed method is measured are sensitivity, specificity, dice coefficient, jacquard index, execution time, accuracy, error rate. From the results, authors have observed proposed model gives the average accuracy value of huge number of cancer images using ISIC dataset is 98.9% and using PH2 dataset is 99.1% with minimize average less error rate. It also estimates the dice coefficient value 0.993 using ISIC and 0.998 using PH2 datasets. However, the results for the rest of the parameters remain quite the same. Therefore the outcome of this model is highly reassuring.

Adversarial example defense based on image reconstruction

The rapid development of deep neural networks (DNN) has promoted the widespread application of image recognition, natural language processing, and autonomous driving. However, DNN is vulnerable to adversarial examples, such as an input sample with imperceptible perturbation which can easily invalidate the DNN and even deliberately modify the classification results. Therefore, this article proposes a preprocessing defense framework based on image compression reconstruction to achieve adversarial example defense. Firstly, the defense framework performs pixel depth compression on the input image based on the sensitivity of the adversarial example to eliminate adversarial perturbations. Secondly, we use the super-resolution image reconstruction network to restore the image quality and then map the adversarial example to the clean image. Therefore, there is no need to modify the network structure of the classifier model, and it can be easily combined with other defense methods. Finally, we evaluate the algorithm with MNIST, Fashion-MNIST, and CIFAR-10 datasets; the experimental results show that our approach outperforms current techniques in the task of defending against adversarial example attacks.

Nanotrap Particles Improve Nanopore Sequencing of SARS-CoV-2 and Other Respiratory Viruses

Presented here is a magnetic hydrogel particle enabled workflow for capturing and concentrating SARS-CoV-2 from diagnostic remnant swab samples that significantly improves sequencing results using the Oxford Nanopore Technologies MinION sequencing platform. Our approach utilizes a novel affinity-based magnetic hydrogel particle, circumventing low input sample volumes and allowing for both rapid manual and automated high throughput workflows that are compatible with nanopore sequencing. This approach enhances standard RNA extraction protocols, providing up to 40x improvements in viral mapped reads, and improves sequencing coverage by 20-80% from lower titer diagnostic remnant samples. Furthermore, we demonstrate that this approach works for contrived influenza virus and respiratory syncytial virus samples, suggesting that it can be used to identify and improve sequencing results of multiple viruses in VTM samples. These methods can be performed manually or on a KingFisher Apex system.

PerSVade: Personalized Structural Variation detection in your species of interest

Structural variants (SVs) like translocations, deletions, and other rearrangements underlie genetic and phenotypic variation. SVs are often overlooked due to difficult detection from short-read sequencing. Most algorithms yield low recall on humans, but the performance in other organisms is unclear. Similarly, despite remarkable differences across species genomes, most approaches use parameters optimized for humans. To overcome this and enable species-tailored approaches, we developed perSVade (personalized Structural Variation Detection), a pipeline that identifies SVs in a way that is optimized for any input sample. Starting from short reads, perSVade uses simulations on the reference genome to choose the best SV calling parameters. The output includes the optimally-called SVs and the accuracy, useful to assess the confidence in the results. In addition, perSVade can call small variants and copy-number variations. In summary, perSVade automatically identifies several types of genomic variation from short reads using sample-optimized parameters. We validated that perSVade increases the SV calling accuracy on simulated variants for six diverse eukaryotes, and on datasets of validated human variants. Importantly, we found no universal set of optimal parameters, which underscores the need for species-specific parameter optimization. PerSVade will improve our understanding about the role of SVs in non-human organisms.

Techniques to Produce and Evaluate Realistic Multivariate Synthetic Data

Background: Proper data modeling in biomedical research requires sufficient data for exploration and reproducibility purposes. A limited sample size can inhibit objective performance evaluation. Objective: We are developing a synthetic population (SP) generation technique to address the limited sample size condition. We show how to estimate a multivariate empirical probability density function (pdf) by converting the task to multiple one-dimensional (1D) pdf estimations. Methods: Kernel density estimation (KDE) in 1D was used to construct univariate maps that converted the input variables (X) to normally distributed variables (Y). Principal component analysis (PCA) was used to transform the variables in Y to the uncoupled representation (T), where the univariate pdfs were assumed normal with specified variances. A standard random number generator was used to create synthetic variables with specified variances in T. Applying the inverse PCA transform to the synthetic variables in T produced the SP in Y. Applying the inverse maps produced the respective SP in X. Multiple tests were developed to compare univariate and multivariate pdfs and covariance matrices between the input (sample) and synthetic samples. Three datasets were investigated (n = 667) each with 10 input variables. Results: For all three datasets, both the univariate (in X, Y, and T) and multivariate (in X, Y, and T) tests showed that the univariate and multivariate pdfs from synthetic samples were statistically similar to their pdfs from the respective samples. Application of several tests for multivariate normality indicated that the SPs in Y were approximately normal. Covariance matrix comparisons (in X and Y) also indicated the same similarity. Conclusions: The work demonstrates how to generate multivariate synthetic data that matches the real input data by converting the input into multiple 1D problems. The work also shows that it is possible to convert a multivariate input pdf to a form that approximates a multivariate normal, although the technique is not dependent upon this finding. Further studies are required to evaluate the generalizability of the approach.

Real-Time Prediction of the Trend of Ground Motion Intensity Based on Deep Learning

In order to predict the intensity of earthquake damage in advance and improve the effectiveness of earthquake emergency measures, this paper proposes a deep learning model for real-time prediction of the trend of ground motion intensity. The input sample is the real-time monitoring recordings of the current received ground motion acceleration. According to the different sampling frequencies, the neural network is constructed by several subnetworks, and the output of each subnetwork is combined into one. After the training and verification of the model, the results show that the model has an accuracy rate of 75% on the testing set, which is effective on real-time prediction of the ground motion intensity. Moreover, the correlation between the Arias intensity and structural damage is stronger than the correlation between peak acceleration and structural damage, so the model is useful for determining real-time response measures on earthquake disaster prevention and mitigation compared with the current more common antiseismic measures based on predictive PGA.

CAMPAREE: a robust and configurable RNA expression simulator

Background The accurate interpretation of RNA-Seq data presents a moving target as scientists continue to introduce new experimental techniques and analysis algorithms. Simulated datasets are an invaluable tool to accurately assess the performance of RNA-Seq analysis methods. However, existing RNA-Seq simulators focus on modeling the technical biases and artifacts of sequencing, rather than on simulating the original RNA samples. A first step in simulating RNA-Seq is to simulate RNA. Results To fill this need, we developed the Configurable And Modular Program Allowing RNA Expression Emulation (CAMPAREE), a simulator using empirical data to simulate diploid RNA samples at the level of individual molecules. We demonstrated CAMPAREE’s use for generating idealized coverage plots from real data, and for adding the ability to generate allele-specific data to existing RNA-Seq simulators that do not natively support this feature. Conclusions Separating input sample modeling from library preparation/sequencing offers added flexibility for both users and developers to mix-and-match different sample and sequencing simulators to suit their specific needs. Furthermore, the ability to maintain sample and sequencing simulators independently provides greater agility to incorporate new biological findings about transcriptomics and new developments in sequencing technologies. Additionally, by simulating at the level of individual molecules, CAMPAREE has the potential to model molecules transcribed from the same genes as a heterogeneous population of transcripts with different states of degradation and processing (splicing, editing, etc.). CAMPAREE was developed in Python, is open source, and freely available at https://github.com/itmat/CAMPAREE.

Learning in Convolutional Neural Networks Accelerated by Transfer Entropy

Recently, there is a growing interest in applying Transfer Entropy (TE) in quantifying the effective connectivity between artificial neurons. In a feedforward network, the TE can be used to quantify the relationships between neuron output pairs located in different layers. Our focus is on how to include the TE in the learning mechanisms of a Convolutional Neural Network (CNN) architecture. We introduce a novel training mechanism for CNN architectures which integrates the TE feedback connections. Adding the TE feedback parameter accelerates the training process, as fewer epochs are needed. On the flip side, it adds computational overhead to each epoch. According to our experiments on CNN classifiers, to achieve a reasonable computational overhead–accuracy trade-off, it is efficient to consider only the inter-neural information transfer of the neuron pairs between the last two fully connected layers. The TE acts as a smoothing factor, generating stability and becoming active only periodically, not after processing each input sample. Therefore, we can consider the TE is in our model a slowly changing meta-parameter.

Modifications of the Multi-Layer Perceptron for Hyperspectral Image Classification

Recently, many convolutional neural network (CNN)-based methods have been proposed to tackle the classification task of hyperspectral images (HSI). In fact, CNN has become the de-facto standard for HSI classification. It seems that the traditional neural networks such as multi-layer perceptron (MLP) are not competitive for HSI classification. However, in this study, we try to prove that the MLP can achieve good classification performance of HSI if it is properly designed and improved. The proposed Modified-MLP for HSI classification contains two special parts: spectral–spatial feature mapping and spectral–spatial information mixing. Specifically, for spectral–spatial feature mapping, each input sample of HSI is divided into a sequence of 3D patches with fixed length and then a linear layer is used to map the 3D patches to spectral–spatial features. For spectral–spatial information mixing, all the spectral–spatial features within a single sample are feed into the solely MLP architecture to model the spectral–spatial information across patches for following HSI classification. Furthermore, to obtain the abundant spectral–spatial information with different scales, Multiscale-MLP is proposed to aggregate neighboring patches with multiscale shapes for acquiring abundant spectral–spatial information. In addition, the Soft-MLP is proposed to further enhance the classification performance by applying soft split operation, which flexibly capture the global relations of patches at different positions in the input HSI sample. Finally, label smoothing is introduced to mitigate the overfitting problem in the Soft-MLP (Soft-MLP-L), which greatly improves the classification performance of MLP-based method. The proposed Modified-MLP, Multiscale-MLP, Soft-MLP, and Soft-MLP-L are tested on the three widely used hyperspectral datasets. The proposed Soft-MLP-L leads to the highest OA, which outperforms CNN by 5.76%, 2.55%, and 2.5% on the Salinas, Pavia, and Indian Pines datasets, respectively. The proposed Modified-MLP, Multiscale-MLP, and Soft-MLP are tested on the three widely used hyperspectral datasets (i.e., Salinas, Pavia, and Indian Pines). The obtained results reveal that the proposed models provide competitive results compared to the state-of-the-art methods, which shows that the MLP-based methods are still competitive for HSI classification.