Graph-Based Fusion of Imaging and Non-Imaging Data for Disease Trajectory Prediction

This study proposes a graph convolutional neural networks (GCN) architecture for fusion of radiological imaging and non-imaging tabular electronic health records (EHR) for the purpose of clinical event prediction. We focused on a cohort of hospitalized patients with positive RT-PCR test for COVID-19 and developed GCN based models to predict three dependent clinical events (discharge from hospital, admission into ICU, and mortality) using demographics, billing codes for procedures and diagnoses and chest X-rays. We hypothesized that the two-fold learning opportunity provided by the GCN is ideal for fusion of imaging information and tabular data as node and edge features, respectively. Our experiments indicate the validity of our hypothesis where GCN based predictive models outperform single modality and traditional fusion models. We compared the proposed models against two variations of imaging-based models, including DenseNet-121 architecture with learnable classification layers and Random Forest classifiers using disease severity score estimated by pre-trained convolutional neural network. GCN based model outperforms both imaging-only methods. We also validated our models on an external dataset where GCN showed valuable generalization capabilities. We noticed that edge-formation function can be adapted even after training the GCN model without limiting application scope of the model. Our models take advantage of this fact for generalization to external data. © 2023 IEEE.

Detection of Semantic Duplicates in Temporal Domains Using Directed Acyclic Graphs

We demonstrate the feasibility of a generalized technique for semantic deduplication in temporal data domains using graph-based representations of data records. Structured data records with multiple timestamp attributes per record may be represented as a directed graph where the nodes represent the events and the edges represent event sequences. Edge weights are based on elapsed time between connecting nodes. In comparing two records, we may merge these directed graphs and determine a representative directed acyclic graph (DAG) inclusive of a subset of nodes and edges that maintain the transitive weights of the original graphs. This DAG may then be evaluated by weighting elapsed time equivalences between records at each node and measuring the fraction of nodes represented in the DAG versus the union of nodes between the records being compared. With this information, we establish a duplication score and use a specified threshold requirement to assert duplication. This method is referred to as Temporal Deduplication using Directed Acyclic Graphs (TD:DAG). TD:DAG significantly outperformed established ASNM and ASNM+LCS methods for datasets rep-resenting two disparate domains, COVID-19 government policy data and PlayStation Network (PSN) trophy data. TD:DAG produced highly effective and comparable F1 scores of 0.960 and 0.972 for the two datasets, respectively, versus 0.864/0.938 for ASNM+LCS and 0.817/0.708 for ASNM. © 2023 IEEE.

Analysis of mobility network metrics in the spread of COVID-19 in the Metropolitan Region of Vale do Paraíba and Litoral Norte (RMVPLN)

In 2019, a pandemic of the so-called new coronavirus (SARS-COV-II) began, which causes the disease COVID-19. In a short time after the first case appeared, hundreds of countries began to register new cases every day. Mapping and analyzing the flow of people, regardless of the mode of transport, can help us to understand and prevent several phenomena that can affect our society in different ways. Graphs are complex networks made up of points and edges. The (geo)graphs are graphs with known spatial location and, in the case of our study, the edges represent the flow between them. The (geo)graphs proved to be a promising tool for such analyses. In the study region, municipalities that first registered their COVID-19 cases are also municipalities that have the highest mobility indices analyzed: degree, betweenness and weight of edges. © 2022 National Institute for Space Research, INPE. All rights reserved.

Study of Deep Learning Approaches for Diagnosing Covid-19 Disease using Chest CT Images

Since March 2020, the World Health Organization (WHO) has declared COVID-19 a pandemic. An evolving viral infection with respiratory tropism causes atypical pneumonia. Experts believe that detecting COVID-19 early stage is crucial. Early diagnosis and tracking techniques have become increasingly important to ensure an accelerated treatment process and avoid virus spread. Images from Computed Tomography (CT) scans can provide quick and precise COVID-19 screening. A subdivision of Machine Learning (ML) called Deep Learning (DL) can improve diagnostic accuracy and speed by automating screening via medical imaging in collaborative efforts with radiologists and physicians This study aims to investigate the recently popularized and extensively discussed deep learning algorithms for COVID-19 diagnosis in connection to the sequence phases involved in image processing. Getting rid of the noise in these images requires some preprocessing. Histogram equalization, fuzzy histogram equalisation, Adaptive Histogram Equalization (AHE) and Contrast Limited Adaptive Histogram Equalization (CLAHE) are used to improve the image quality and therefore increase the identification of the image. Afterwards, necessary features for disease detection are segmented using various deep models like U-Net, U-Net + FPN (Feature Pyramid Network), COVID-SegNet and Dense GAN. Once these distinct deep characteristics have been identified, they are extracted using a variety of different deep models. Finally, an illness is diagnosed using popular models such as SVM, ResNet-50, AlexNet, VGG16, DenseNet, and SqueezeNet. The deep learning models with a better optimization algorithm to be effective in the diagnosis of COVID-19 and also obtain a reduced and efficient feature set for image classification and feature extraction. © 2023 IEEE.

Inductive Graph-based Knowledge Tracing

The rise of virtual education and increase in distance, partly owing to the spread of COVID-19 pandemic, has made it more difficult for teachers to determine each student's learning status. In this situation, knowledge tracing (KT), which tracks a student's mastery of specific knowledge concepts, is receiving increasing attention. KT utilizes a sequence of studentexercise interactive activities to predict the mastery of concepts corresponding to a target problem, recommending appropriate learning resources to students and optimizing learning sequences for adaptive learning. With the development of deep learning, various studies have been proposed, such as sequential models using recurrent neural networks, attention models influenced by transformers, and graph-based models that depict the relationships between knowledge concepts. However, they all have common limitations in that they cannot utilize the learning activities of students other than the target student and can only use a limited form of exercise information. In this study, we have applied the concept of rating prediction to the studentexercise knowledge tracing problem and solved the limitations of the existing models. Our proposed Inductive Graph-based Knowledge Tracing (IGKT) designed to integrate structural information and various unrestricted types of additional information into the model through subgraph sampling, has been found superior over the existing models across two different datasets in predicting student performances. © 2023 IEEE.

Identification of COVID-19 in CT Images using Deep Learning Model

Worldwide, COVID-19 has had a substantial impact on patients and hospital systems. Early identification and diagnosis are essential for regulating the growth of COVID-19. The input CT screening images are initially segmented into various regions using the Fuzzy C-means (FCM) clustering technique. Next, region-based image quality enhancement employs a histogram equalization method. Furthermore, certain necessary data is represented in a new image using the Local Directional Number technique. Lastly, the input images are portioned with the help of a traditional convolutional neural network model. The proposed convolutional neural network based system was able to give an accuracy of 98.60%, and the results revealed that methods for detecting COVID-19 impact from CT scan images must be developed significantly before considering it as a medical choice. Moreover, many diverse datasets are essential to assess the processes in a real-world setting. © 2022 IEEE.

An End-to-End Knowledge Graph Based Question Answering Approach for COVID-19

Question Answering based on Knowledge Graph (KG) has emerged as a popular research area in general domain. However, few works focus on the COVID-19 kg-based question answering, which is very valuable for biomedical domain. In addition, existing question answering methods rely on knowledge embedding models to represent knowledge (i.e., entities and questions), but the relations between entities are neglected. In this paper, we construct a COVID-19 knowledge graph and propose an end-to-end knowledge graph question answering approach that can utilize relation information to improve the performance. Experimental result shows that the effectiveness of our approach on the COVID-19 knowledge graph question answering. Our code and data are available at https://github.com/CHNcreater/COVID-19-KGQA. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Smart Attendance System using Dlib Pre-Trained Neural Network

Recognizing and remembering various people is the most frequent task, which the human brain performs. With regard to this, the process of attendance becomes one of the hectic tasks, which requires subsequent modernization. The spread of COVID- 19 is also drastically increasing and are pushed to the situation of wearing mask the entire time. This brings in a situation of misidentifying the individuals and are also prone to impersonation in many official gatherings such as exams, meetings, etc. This cannot be decreased by unmasking their face in this pandemic situation just for the purpose of verification as it may lead to increase in COVID risk. Here, this research study implements a contactless face recognition system with a simple and smart database, which can take in any form of data as per the convenience. This system solves the above problem by making the face recognition smart using Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) classifier. The main task of the system is to recognize the user's face (live) and automatically mark the time of recognition directly in the Google sheet along with the alphabets of P(Present), A(absent) or L(late) according to the given time range. This system makes effective use of google sheet for easy share ability, accessibility, and error free management. This can be used for number of purposes such as exam centers, schools, colleges, companies, hospitals and various other places in order to verify the people (contact less). © 2023 IEEE.

Graggle: A Graph-based Approach to Document Clustering

Document recommendation systems have traditionally relied upon high-dimensional vector representations that scale poorly in corpora with diverse vocabularies. Existing graph-based approaches focus on the metadata of documents and, unfortunately, ignore the content of the papers. In this work, we have designed and implemented a new system we call Graggle, which builds a graph to model a corpus. Nodes are papers, and edges represent significant words shared between them. We then leverage modern graph learning techniques to turn this graph into a highly efficient tool for dimensionality reduction. Documents are represented as low-dimensional vector embeddings generated with a graph autoencoder. Our experiments show that this approach outperforms traditional document vector-based and text autoencoding approaches on labeled data. Additionally, we have applied this technique to a repository of unlabeled research documents about the novel coronavirus to demonstrate its effectiveness as a real-world tool. © 2022 IEEE.

A graph-based edge attention gate medical image segmentation method

For the purpose of solving the problems of missing edges and low segmentation accuracy in medical image segmentation, a medical image segmentation network (EAGC_UNet++) based on residual graph convolution UNet++ with edge attention gate (EAG) is proposed in the study. With UNet++ as the backbone network, the idea of graph theory is introduced into the model. First, the dropout residual graph convolution block (DropRes_GCN Block) and the traditional convolution structure in UNet++ are used as encoders. Second, EAGs are adopted so that the model pays more attention to image edge features during decoding. Finally, aiming at the imbalance problem of positive and negative samples in medical image segmentation, a new weighted loss function is introduced to enhance segmentation accuracy. In the experimental part, three datasets (LiTS2017, ISIC2018, COVID-19 CT scans) were used to evaluate the performances of various models;multiple groups of ablation experiments were designed to verify the effectiveness of each part of the model. The experimental results showed that EAGC_UNet++ had better segmentation performance than the other models under three quantitative evaluation indicators and better solved the problem of missing edges in medical image segmentation. © 2023 The Authors. IET Image Processing published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

HateNet: A Graph Convolutional Network Approach to Hate Speech Detection

The COVID-19 pandemic has caused hate speech on online social networks to become a growing issue in recent years, affecting millions. Our work aims to improve automatic hate speech detection to prevent escalation to hate crimes. The first c hallenge i n h ate s peech r esearch i s t hat e xisting datasets suffer from quite severe class imbalances. The second challenge is the sparsity of information in textual data. The third challenge is the difficulty i n b alancing t he t radeoff b etween utilizing semantic similarity and noisy network language. To combat these challenges, we establish a framework for automatic short text data augmentation by using a semi-supervised hybrid of Substitution Based Augmentation and Dynamic Query Expansion (DQE), which we refer to as SubDQE, to extract more data points from a specific c lass f rom T witter. W e a lso p ropose the HateNet model, which has two main components, a Graph Convolutional Network and a Weighted Drop-Edge. First, we propose a Graph Convolutional Network (GCN) classifier, using a graph constructed from the thresholded cosine similarities between tweet embeddings to provide new insights into how ideas are connected. Second, we propose a weighted Drop-Edge based stochastic regularization technique, which removes edges randomly based on weighted probabilities assigned by the semantic similarities between Tweets. Using 3 different SubDQE-augmented datasets, we compare our HateNet model using eight different tweet embedding methods, six other baseline classification models, and seven other baseline data augmentation techniques previously used in the realm of hate speech detection. Our results show that our proposed HateNet model matches or exceeds the performance of the baseline models, as indicated by the accuracy and F1 score. © 2022 IEEE.

Few-Shot Semantic Relation Prediction Across Heterogeneous Graphs

Semantic relation prediction aims to mine the implicit relationships between objects in heterogeneous graphs, which consist of different types of objects and different types of links. In real-world scenarios, new semantic relations constantly emerge and they typically appear with only a few labeled data. Since a variety of semantic relations exist in multiple heterogeneous graphs, the transferable knowledge can be mined from some existing semantic relations to help predict the new semantic relations with few labeled data. This inspires a novel problem of few-shot semantic relation prediction across heterogeneous graphs. However, the existing methods cannot solve this problem because they not only require a large number of labeled samples as input, but also focus on a single graph with a fixed heterogeneity. Targeting this novel and challenging problem, in this paper, we propose a Meta-learning based Graph neural network for Semantic relation prediction, named MetaGS. Firstly, MetaGS decomposes the graph structure between objects into multiple normalized subgraphs, then adopts a two-view graph neural network to capture local heterogeneous information and global structure information of these subgraphs. Secondly, MetaGS aggregates the information of these subgraphs with a hyper-prototypical network, which can learn from existing semantic relations and adapt to new semantic relations. Thirdly, using the well-initialized two-view graph neural network and hyper-prototypical network, MetaGS can effectively learn new semantic relations from different graphs while overcoming the limitation of few labeled data. Extensive experiments on three real-world datasets have demonstrated the superior performance of MetaGS over the state-of-the-art methods. IEEE

Deep Graph Networks for Drug Repurposing with Multi-Protein Targets

In the early phases of the COVID-19 pandemic, repurposing of drugs approved for use in other diseases helped counteract the aggressiveness of the virus. Therefore, the availability of effective and flexible methodologies to speed up and prioritize the repurposing process is fundamental to tackle present and future challenges to worldwide health. This work addresses the problem of drug repurposing through the lens of deep learning for graphs, by designing an architecture that exploits both structural and biological information to propose a reduced set of drugs that may be effective against an unknown disease. Our main contribution is a method to repurpose a drug against multiple proteins, rather than the most common single-drug/single-protein setting. The method leverages graph embeddings to encode the relevant proteins'and drugs'information based on gene ontology data and structural similarities. Finally, we publicly release a comprehensive and unified data repository for graph-based analysis to foster further studies on COVID-19 and drug repurposing. We empirically validate the proposed approach in a general drug repurposing setting, showing that it generalizes better than single protein repurposing schemes. We conclude the manuscript with an exemplified application of our method to the COVID-19 use case. All source code is publicly available. IEEE

SPATIAL GRAPH ATTENTION AND CURIOSITY-DRIVEN POLICY FOR ANTIVIRAL DRUG DISCOVERY

We developed Distilled Graph Attention Policy Network (DGAPN), a reinforcement learning model to generate novel graph-structured chemical representations that optimize user-defined objectives by efficiently navigating a physically constrained domain. The framework is examined on the task of generating molecules that are designed to bind, noncovalently, to functional sites of SARS-CoV-2 proteins. We present a spatial Graph Attention (sGAT) mechanism that leverages self-attention over both node and edge attributes as well as encoding the spatial structure - this capability is of considerable interest in synthetic biology and drug discovery. An attentional policy network is introduced to learn the decision rules for a dynamic, fragment-based chemical environment, and state-of-the-art policy gradient techniques are employed to train the network with stability. Exploration is driven by the stochasticity of the action space design and the innovation reward bonuses learned and proposed by random network distillation. In experiments, our framework achieved outstanding results compared to state-of-the-art algorithms, while reducing the complexity of paths to chemical synthesis. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

Face Mask Detection Using Viola-Jones and Cascade Classifier

Globally, the COVID-19 coronavirus outbreak is causing chaos in human health and therefore, the healthcare sector is in serious disarray. Many precautions have been taken to prevent the spread of this disease, including the usage of masks, which is strongly recommended by the World Health Organization (WHO). This research study has used the Viola-Jones algorithm for detecting face masks, where Histogram Equalization, Unsharp Filter and Gamma Correction are used as the preferred image pre-processing techniques to improve the overall accuracy. Haar Feature Selection is applied for creating integral images and AdaBoost training is performed on these images. Cascade classifier, a machine learning-based approach, is also integrated with the base algorithm where a cascade function assists Viola-Jones in accurately detecting objects in images. A total number of 1670 images is used in this work and our system is compared with four other machine learning algorithms, where Viola-Jones outperforms these ML-based classifiers and the overall accuracy obtained is 96%. © 2022 IEEE.

A Real-Time Crowd Detection and Monitoring System using Machine Learning

The COVID-19 pandemic has unquestionably warned all of us that, the outbreak of an infection can lead to a pandemic-like situation all over the world. In order to prevent outbreaks and provide better healthcare, appropriate crowd detection and monitoring systems must be deployed in public areas. By effectively implementing social distancing measures, the number of new infections can be greatly decreased. This idea served as the inspiration for the creation of a real-time Crowd Detection and Monitoring System (CDMS) for social distancing. This paper proposes a fully autonomous system for Real-Time Crowd Detection and Monitoring to help the educational institutions to monitor the students inside the premises more effectively. This system is developed using an OpenCV based Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) detector to detect and count the number of people gathered at an instance. The system raises an alarm to alert the people and adhere to the rules if the gathering is more than the threshold/permitted number of people in the cluster. © 2023 IEEE.

Unsupervised Identification of Abnormal Nodes and Edges in Graphs

Much of today's data are represented as graphs, ranging from social networks to bibliographic citations. Nodes in such graphs correspond to records that generally represent entities, while edges represent relationships between these entities. Both nodes and edges in a graph can have attributes that characterize the entities and their relationships. Relationships are either explicitly known (like friends in a social network), or they are inferred using link prediction (such as two babies are siblings because they have the same mother). Any graph representing real-world data likely contains nodes and edges that are abnormal, and identifying these can be important for outlier detection in applications ranging from crime and fraud detection to viral marketing. We propose a novel approach to the unsupervised detection of abnormal nodes and edges in graphs. We first characterize nodes and edges using a set of features, and then employ a one-class classifier to identify abnormal nodes and edges. We extract patterns of features from these abnormal nodes and edges, and apply clustering to identify groups of patterns with similar characteristics. We finally visualize these abnormal patterns to show co-occurrences of features and relationships between those features that mostly influence the abnormality of nodes and edges. We evaluate our approach on datasets from diverse domains, including historical birth certificates, COVID patient records, e-mails, books, and movies. This evaluation demonstrates that our approach is well suited to identify both abnormal nodes and edges in graphs in an unsupervised way, and it can outperform several baseline anomaly detection techniques. © 2022 Copyright held by the owner/author(s).

A high-speed unsupervised hardware architecture for rapid diagnosis of COVID-19

In the diagnosis of COVID-19, investigation, analysis, and automatic counting of blood cell clusters are the most essential steps. Currently employed methods for cell segmentation, identification, and counting are time-consuming and sometimes performed manually from sampled blood smears, which is hard and needs the support of an expert laboratory technician. The conventional method for the blood-count-test is by automatic hematology analyzer which is quite expensive and slow. Moreover, most of the unsupervised learning techniques currently available presume the medical practitioner to have a prior knowledge regarding the number and action of possible segments within the image before applying recognition. This assumption fails most often as the severity of the disease gets increased like the advanced stages of COVID-19, lung cancer etc. In this manuscript, a simplified automatic histopathological image analysis technique and its hardware architecture suited for blind segmentation, cell counting, and retrieving the cell parameters like radii, area, and perimeter has been identified not only to speed up but also to ease the process of diagnosis as well as prognosis of COVID-19. This is achieved by combining three algorithms: the K-means algorithm, a novel statistical analysis technique-HIST (histogram separation technique), and an islanding method an improved version of CCA algorithm/blob detection technique. The proposed method is applied to 15 chronic respiratory disease cases of COVID-19 taken from high profile hospital databases. The output in terms of quantitative parameters like PSNR, SSIM, and qualitative analysis clearly reveals the usefulness of this technique in quick cytological evaluation. The proposed high-speed and low-cost architecture gives promising results in terms of performance of 190 MHz clock frequency, which is two times faster than its software implementation. © 2022 John Wiley & Sons Ltd.

Lung disease detection using Self-Attention Generative Adversarial Capsule network optimized with sun flower Optimization Algorithm

Lung cancer is the uncontrolled growth of abnormal cells in one or both lungs. This is one of the dangerous diseases. A lot of feature extraction with classification methods were discussed previously regarding this disease, but none of the methods give sufficient results, not only that, those methods have high over fitting problem, as a result, the detection accuracy was minimizing. Therefore, to overcome these issues, a Lung Disease Detection using Self-Attention Generative Adversarial Capsule Network optimized with Sun flower Optimization Algorithm (SA-Caps GAN-SFOA-LDC) is proposed in this manuscript. Initially, NIH chest X-ray image dataset is gathered through Kaggle repository to diagnose the lung disease. Then, the chests X-ray images are pre-processed by using the contrast limited adaptive histogram equalization (CLAHE) filtering method to eliminate the noise and to enhance the image quality. These pre-processed outputs are fed to feature extraction process. In the feature extraction process, the empirical wavelet transform method is used. These extracted features are given into Self-Attention based Generative Adversarial Capsule classifier for detecting the lung disease. The hyper parameters of SA-Caps GAN classifier is optimized using Sun flower Optimization Algorithm. The simulation is implemented in MATLAB. The proposed SA-Caps GAN-SFOA-LDC method attains higher accuracy 21.05%, 33.28%, 30.27%, 29.68%, 32.57% and 44.28%, Higher Precision 30.24%, 35.68%, 32.08%, 41.27%, 28.57% and 34.20%, Higher F-Score 32.05%, 31.05%, 36.24%, 30.27%, 37.59% and 22.05% analyzed with the existing methods, SVM-SMO-LDC, CNN-MOSHO-LDC, XGboost-PSO-LDC respectively. © 2022 Elsevier Ltd

Twitter use by the dementia community during COVID-19: a user classification and social network analysis

Purpose: The study aimed to examine how different communities concerned with dementia engage and interact on Twitter. Design/methodology/approach: A dataset was sampled from 8,400 user profile descriptions, which was labelled into five categories and subjected to multiple machine learning (ML) classification experiments based on text features to classify user categories. Social network analysis (SNA) was used to identify influential communities via graph-based metrics on user categories. The relationship between bot score and network metrics in these groups was also explored. Findings: Classification accuracy values were achieved at 82% using support vector machine (SVM). The SNA revealed influential behaviour on both the category and node levels. About 2.19% suspected social bots contributed to the coronavirus disease 2019 (COVID-19) dementia discussions in different communities. Originality/value: The study is a unique attempt to apply SNA to examine the most influential groups of Twitter users in the dementia community. The findings also highlight the capability of ML methods for efficient multi-category classification in a crisis, considering the fast-paced generation of data. Peer review: The peer review history for this article is available at: https://publons.com/publon/10.1108/OIR-04-2021-0208. © 2022, Emerald Publishing Limited.