Early Detection of Hemorrhagic Stroke Using a Lightweight Deep Learning Neural Network Model

In present days, the major disease affecting people all across the world is “Cerebrovascular Stroke”. Computed tomographic (CT) images play a crucial role in identifying hemorrhagic strokes. It also helps in understanding the impact of damage caused in the brain cells more accurately. The existing research work is implemented on the Graphical Processing Unit (GPU’s) for stroke segmentation using heavyweight convolutions that require more processing time for diagnosis and increases the model's cost. Deep learning techniques with VGG-16 architecture and Random Forest algorithm are implemented for detecting hemorrhagic stroke using brain CT images under segmentation. A two-step light-weighted convolution model is proposed by using the data collected from multiple- repositories to inscribe this constraint. In the first step, the input CT images are given to VGG-16 architecture and in step two, data frames are given to random forest for stroke segmentation with three levels of classes. In this paper, we explore various training time values in the detection of stroke that reduces when compared with existing heavyweight models. Experimental results have shown that when compared to other existing architectures, our hybrid model VGG-16 and random forest achieved increased results obtained are dice coefficient with 72.92 and accuracy with 97.81% which shows promising results.

Identifying Cancer Subtypes Using a Residual Graph Convolution Model on a Sample Similarity Network

Cancer subtype classification helps us to understand the pathogenesis of cancer and develop new cancer drugs, treatment from which patients would benefit most. Most previous studies detect cancer subtypes by extracting features from individual samples, ignoring their associations with others. We believe that the interactions of cancer samples can help identify cancer subtypes. This work proposes a cancer subtype classification method based on a residual graph convolutional network and a sample similarity network. First, we constructed a sample similarity network regarding cancer gene co-expression patterns. Then, the gene expression profiles of cancer samples as initial features and the sample similarity network were passed into a two-layer graph convolutional network (GCN) model. We introduced the initial features to the GCN model to avoid over-smoothing during the training process. Finally, the classification of cancer subtypes was obtained through a softmax activation function. Our model was applied to breast invasive carcinoma (BRCA), glioblastoma multiforme (GBM) and lung cancer (LUNG) datasets. The accuracy values of our model reached 82.58%, 85.13% and 79.18% for BRCA, GBM and LUNG, respectively, which outperformed the existing methods. The survival analysis of our results proves the significant clinical features of the cancer subtypes identified by our model. Moreover, we can leverage our model to detect the essential genes enriched in gene ontology (GO) terms and the biological pathways related to a cancer subtype.

Estimation of the depth-domain seismic wavelet based on velocity substitution and a generalized seismic wavelet model

Depth-domain seismic wavelet estimation is the essential foundation for depth-imaged data inversion, which is increasingly used for hydrocarbon reservoir characterization in geophysical prospecting. The seismic wavelet in the depth domain stretches with the medium velocity increase and compresses with the medium velocity decrease. The commonly used convolution model cannot be directly used to estimate depth-domain seismic wavelets due to velocity-dependent wavelet variations. We develop a separate parameter estimation method for estimating depth-domain seismic wavelets from poststack depth-domain seismic and well log data. This method is based on the velocity substitution and depth-domain generalized seismic wavelet model defined by the fractional derivative and reference wavenumber. Velocity substitution allows wavelet estimation with the convolution model in the constant-velocity depth domain. The depth-domain generalized seismic wavelet model allows for a simple workflow that estimates the depth-domain wavelet by estimating two wavelet model parameters. Additionally, this simple workflow does not need to perform searches for the optimal regularization parameter and wavelet length, which are time-consuming in least-squares-based methods. The limited numerical search ranges of the two wavelet model parameters can easily be calculated using the constant phase and peak wavenumber of the depth-domain seismic data. Our method is verified using synthetic and real seismic data and further compared with least-squares-based methods. The results indicate that the proposed method is effective and stable even for data with a low S/N.

A Spectrogram Based Novel Approach for Arrhythmia Detection with Convolutional Neural Networks

ECG is one of the most important medical scans which is used for diagnosis of various heart related conditions and diseases. One of the most common of these is arrhythmia, which is caused by the irregularity of the heart beats. Artificial Intelligence has had a major impact in the field of vital monitoring and autonomous medical diagnosis. Therefore, a lot of work has demonstrated its effectiveness in arrhythmia detection. In this paper, we propose a method that tries to improve upon the accuracy of such models with the help of a light weight deep learning architecture that utilized 2D Separable CNN with a group of graphical representations of the ECG signals like the STFT, CWT and MFCC. Our model has achieved an accuracy of 97.41 and an F1 score of 88.20 on a processed version of the MIT-BIH dataset and takes on an average 7.93 times less calculations compared to a simple 2D Convolution model.

Convolution model for COVID-19 rate predictions and health effort levels computation for Saudi Arabia, France, and Canada

Many published infection prediction models, such as the extended SEIR (E-SEIR) model, are used as a study and report tool to aid health authorities to manage the epidemic plans successfully. These models face many challenges, mainly the reliability of the infection rate predictions related to the initial boundary conditions, formulation complexity, lengthy computations, and the limited result scope. We attribute these challenges to the absence of a solution framework that encapsulates the interacted activities that manage: the infection growth process, the infection spread process and the health effort process. In response to these challenges, we formulated such a framework first as the basis of our new convolution prediction model (CPM). CPM links through convolution integration, three temporal profile levels: input (infected and active cases), transformational (health efforts), and output functions (recovered, quarantine, and death cases). COVID-19 data defines the input and output temporal profiles; hence it is possible to deduce the cumulative efforts temporal response (CETR) function for the health effort level. The new CETR function determines the health effort level over a period. Also, CETR plays a role in predicting the evolution of the underlying infection and active cases profiles without a system of differential equations. This work covers three countries: Saudi Arabia, France, and Canada.

EOESGC: predicting miRNA-disease associations based on embedding of embedding and simplified graph convolutional network

Background A large number of biological studies have shown that miRNAs are inextricably linked to many complex diseases. Studying the miRNA-disease associations could provide us a root cause understanding of the underlying pathogenesis in which promotes the progress of drug development. However, traditional biological experiments are very time-consuming and costly. Therefore, we come up with an efficient models to solve this challenge. Results In this work, we propose a deep learning model called EOESGC to predict potential miRNA-disease associations based on embedding of embedding and simplified convolutional network. Firstly, integrated disease similarity, integrated miRNA similarity, and miRNA-disease association network are used to construct a coupled heterogeneous graph, and the edges with low similarity are removed to simplify the graph structure and ensure the effectiveness of edges. Secondly, the Embedding of embedding model (EOE) is used to learn edge information in the coupled heterogeneous graph. The training rule of the model is that the associated nodes are close to each other and the unassociated nodes are far away from each other. Based on this rule, edge information learned is added into node embedding as supplementary information to enrich node information. Then, node embedding of EOE model training as a new feature of miRNA and disease, and information aggregation is performed by simplified graph convolution model, in which each level of convolution can aggregate multi-hop neighbor information. In this step, we only use the miRNA-disease association network to further simplify the graph structure, thus reducing the computational complexity. Finally, feature embeddings of both miRNA and disease are spliced into the MLP for prediction. On the EOESGC evaluation part, the AUC, AUPR, and F1-score of our model are 0.9658, 0.8543 and 0.8644 by 5-fold cross-validation respectively. Compared with the latest published models, our model shows better results. In addition, we predict the top 20 potential miRNAs for breast cancer and lung cancer, most of which are validated in the dbDEMC and HMDD3.2 databases. Conclusion The comprehensive experimental results show that EOESGC can effectively identify the potential miRNA-disease associations.

Development and prediction of flame retarded particleboard fire behaviour in real scale

Particleboard composites for building application has become very attractive because of their huge benefits which includes but is not limited to low cost, lightweight, durability and environmental benign. However, the vulnerability of these composite types when exposed to fire restricts their use in areas where stringent fire safety conditions may not apply. Experimental determination at bench scale of composite particleboard fire behaviour has shown that the addition of flame retardants (FR) can delay the start and spread of fire. Bench scale data obtained in the cone calorimeter (CC) may not represent accurately a real scale fore behaviour during a fire scenario as documented by various researchers. The convolution model is a significant tool for predicting in real scale, fire behaviour of composites which depends on experimental inputs from CC data. In this paper, particleboards made from wood sawdust reinforced polyester composite were processed with FR at 0, 15 and 18% loading ratio using compression moulding technique. Test specimens cut from the FR-particleboards was exposed in horizontal orientation in the CC at 50kW/m2 to obtain experimental data and these were used as inputs to the prediction model. The predictive tool was used to predict the heat released rate and smoke production rate for the FR-particleboard. The results obtained shows that the FR-particleboard contributes very limited fire in real scale and compares well based on Euro-classification with particleboards from literature.