Predicting preterm birth using explainable machine learning in a prospective cohort of nulliparous and multiparous pregnant women.

Preterm birth (PTB) presents a complex challenge in pregnancy, often leading to significant perinatal and long-term morbidities. "While machine learning (ML) algorithms have shown promise in PTB prediction, the lack of interpretability in existing models hinders their clinical utility. This study aimed to predict PTB in a pregnant population using ML models, identify the key risk factors associated with PTB through the SHapley Additive exPlanations (SHAP) algorithm, and provide comprehensive explanations for these predictions to assist clinicians in providing appropriate care. This study analyzed a dataset of 3509 pregnant women in the United Arab Emirates and selected 35 risk factors associated with PTB based on the existing medical and artificial intelligence literature. Six ML algorithms were tested, wherein the XGBoost model exhibited the best performance, with an area under the operator receiving curves of 0.735 and 0.723 for parous and nulliparous women, respectively. The SHAP feature attribution framework was employed to identify the most significant risk factors linked to PTB. Additionally, individual patient analysis was performed using the SHAP and the local interpretable model-agnostic explanation algorithms (LIME). The overall incidence of PTB was 11.23% (11 and 12.1% in parous and nulliparous women, respectively). The main risk factors associated with PTB in parous women are previous PTB, previous cesarean section, preeclampsia during pregnancy, and maternal age. In nulliparous women, body mass index at delivery, maternal age, and the presence of amniotic infection were the most relevant risk factors. The trained ML prediction model developed in this study holds promise as a valuable screening tool for predicting PTB within this specific population. Furthermore, SHAP and LIME analyses can assist clinicians in understanding the individualized impact of each risk factor on their patients and provide appropriate care to reduce morbidity and mortality related to PTB.

Node embedding-based graph autoencoder outlier detection for adverse pregnancy outcomes.

Adverse pregnancy outcomes, such as low birth weight (LBW) and preterm birth (PTB), can have serious consequences for both the mother and infant. Early prediction of such outcomes is important for their prevention. Previous studies using traditional machine learning (ML) models for predicting PTB and LBW have encountered two important limitations: extreme class imbalance in medical datasets and the inability to account for complex relational structures between entities. To address these limitations, we propose a node embedding-based graph outlier detection algorithm to predict adverse pregnancy outcomes. We developed a knowledge graph using a well-curated representative dataset of the Emirati population and two node embedding algorithms. The graph autoencoder (GAE) was trained by applying a combination of original risk factors and node embedding features. Samples that were difficult to reconstruct at the output of GAE were identified as outliers considered representing PTB and LBW samples. Our experiments using LBW, PTB, and very PTB datasets demonstrated that incorporating node embedding considerably improved performance, achieving a 12% higher AUC-ROC compared to traditional GAE. Our study demonstrates the effectiveness of node embedding and graph outlier detection in improving the prediction performance of adverse pregnancy outcomes in well-curated population datasets.

Deep learning approach to security enforcement in cloud workflow orchestration.

Supporting security and data privacy in cloud workflows has attracted significant research attention. For example, private patients' data managed by a workflow deployed on the cloud need to be protected, and communication of such data across multiple stakeholders should also be secured. In general, security threats in cloud environments have been studied extensively. Such threats include data breaches, data loss, denial of service, service rejection, and malicious insiders generated from issues such as multi-tenancy, loss of control over data and trust. Supporting the security of a cloud workflow deployed and executed over a dynamic environment, across different platforms, involving different stakeholders, and dynamic data is a difficult task and is the sole responsibility of cloud providers. Therefore, in this paper, we propose an architecture and a formal model for security enforcement in cloud workflow orchestration. The proposed architecture emphasizes monitoring cloud resources, workflow tasks, and the data to detect and predict anomalies in cloud workflow orchestration using a multi-modal approach that combines deep learning, one class classification, and clustering. It also features an adaptation scheme to cope with anomalies and mitigate their effect on the workflow cloud performance. Our prediction model captures unsupervised static and dynamic features as well as reduces the data dimensionality, which leads to better characterization of various cloud workflow tasks, and thus provides better prediction of potential attacks. We conduct a set of experiments to evaluate the proposed anomaly detection, prediction, and adaptation schemes using a real COVID-19 dataset of patient health records. The results of the training and prediction experiments show high anomaly prediction accuracy in terms of precision, recall, and F1 scores. Other experimental results maintained a high execution performance of the cloud workflow after applying adaptation strategy to respond to some detected anomalies. The experiments demonstrate how the proposed architecture prevents unnecessary wastage of resources due to anomaly detection and prediction.

Infant birth weight estimation and low birth weight classification in United Arab Emirates using machine learning algorithms.

Accurate prediction of a newborn's birth weight (BW) is a crucial determinant to evaluate the newborn's health and safety. Infants with low BW (LBW) are at a higher risk of serious short- and long-term health outcomes. Over the past decade, machine learning (ML) techniques have shown a successful breakthrough in the field of medical diagnostics. Various automated systems have been proposed that use maternal features for LBW prediction. However, each proposed system uses different maternal features for LBW classification and estimation. Therefore, this paper provides a detailed setup for BW estimation and LBW classification. Multiple subsets of features were combined to perform predictions with and without feature selection techniques. Furthermore, the synthetic minority oversampling technique was employed to oversample the minority class. The performance of 30 ML algorithms was evaluated for both infant BW estimation and LBW classification. Experiments were performed on a self-created dataset with 88 features. The dataset was obtained from 821 women from three hospitals in the United Arab Emirates. Different performance metrics, such as mean absolute error and mean absolute percent error, were used for BW estimation. Accuracy, precision, recall, F-scores, and confusion matrices were used for LBW classification. Extensive experiments performed using five-folds cross validation show that the best weight estimation was obtained using Random Forest algorithm with mean absolute error of 294.53 g while the best classification performance was obtained using Logistic Regression with SMOTE oversampling techniques that achieved accuracy, precision, recall and F1 score of 90.24%, 87.6%, 90.2% and 0.89, respectively. The results also suggest that features such as diabetes, hypertension, and gestational age, play a vital role in LBW classification.

DeeProBot: a hybrid deep neural network model for social bot detection based on user profile data.

Use of online social networks (OSNs) undoubtedly brings the world closer. OSNs like Twitter provide a space for expressing one's opinions in a public platform. This great potential is misused by the creation of bot accounts, which spread fake news and manipulate opinions. Hence, distinguishing genuine human accounts from bot accounts has become a pressing issue for researchers. In this paper, we propose a framework based on deep learning to classify Twitter accounts as either 'human' or 'bot.' We use the information from user profile metadata of the Twitter account like description, follower count and tweet count. We name the framework 'DeeProBot,' which stands for Deep Profile-based Bot detection framework. The raw text from the description field of the Twitter account is also considered a feature for training the model by embedding the raw text using pre-trained Global Vectors (GLoVe) for word representation. Using only the user profile-based features considerably reduces the feature engineering overhead compared with that of user timeline-based features like user tweets and retweets. DeeProBot handles mixed types of features including numerical, binary, and text data, making the model hybrid. The network is designed with long short-term memory (LSTM) units and dense layers to accept and process the mixed input types. The proposed model is evaluated on a collection of publicly available labeled datasets. We have designed the model to make it generalizable across different datasets. The model is evaluated using two ways: testing on a hold-out set of the same dataset; and training with one dataset and testing with a different dataset. With these experiments, the proposed model achieved AUC as high as 0.97 with a selected set of features.

Vaccine versus Variants (3Vs): Are the COVID-19 Vaccines Effective against the Variants? A Systematic Review.

BACKGROUND: With the emergence and spread of new SARS-CoV-2 variants, concerns are raised about the effectiveness of the existing vaccines to protect against these new variants. Although many vaccines were found to be highly effective against the reference COVID-19 strain, the same level of protection may not be found against mutation strains. The objective of this study is to systematically review relevant studies in the literature and compare the efficacy of COVID-19 vaccines against new variants. METHODS: We conducted a systematic review of research published in Scopus, PubMed, and Google Scholar until 30 August 2021. Studies including clinical trials, prospective cohorts, retrospective cohorts, and test negative case-controls that reported vaccine effectiveness against any COVID-19 variants were considered. PRISMA recommendations were adopted for screening, eligibility, and inclusion. RESULTS: 129 unique studies were reviewed by the search criteria, of which 35 met the inclusion criteria. These comprised of 13 test negative case-control studies, 6 Phase 1-3 clinical trials, and 16 observational studies. The study location, type, vaccines used, variants considered, and reported efficacies were highlighted. CONCLUSION: Full vaccination (two doses) offers strong protection against Alpha (B.1.1.7) with 13 out of 15 studies reporting more than 84% efficacy. The results are not conclusive against the Beta (B.1.351) variant for fully vaccinated individuals with 4 out of 7 studies reporting efficacies between 22 and 60% and 3 out of 7 studies reporting efficacies between 75 and 100%. Protection against Gamma (P.1) variant was lower than 50% according to two studies in fully vaccinated individuals. The data on Delta (B.1.617.2) variant is limited but indicates lower protection compared to other variants.

Resource-Aware Mobile-Based Health Monitoring.

Monitoring heart diseases often requires frequent measurements of electrocardiogram (ECG) signals at different periods of the day, and at different situations (e.g., traveling, and exercising). This can only be implemented using mobile devices in order to cope with mobility of patients under monitoring, thus supporting continuous monitoring practices. However, these devices are energy-aware, have limited computing resources (e.g., CPU speed and memory), and might lose network connectivity, which makes it very challenging to maintain a continuity of the monitoring episode. In this paper, we propose a mobile monitoring solution to cope with these challenges by compromising on the fly resources availability, battery level, and network intermittence. In order to solve this problem, first we divide the whole process into several subtasks such that each subtask can be executed sequentially either in the server or in the mobile or in parallel in both devices. Then, we developed a mathematical model that considers all the constraints and finds a dynamic programing solution to obtain the best execution path (i.e., which substep should be done where). The solution guarantees an optimum execution time, while considering device battery availability, execution and transmission time, and network availability. We conducted a series of experiments to evaluate our proposed approach using some key monitoring tasks starting from preprocessing to classification and prediction. The results we have obtained proved that our approach gives the best (lowest) running time for any combination of factors including processing speed, input size, and network bandwidth. Compared to several greedy but nonoptimal solutions, the execution time of our approach was at least 10 times faster and consumed 90% less energy.

Evaluation of individual skeletal muscle activity by glucose uptake during pedaling exercise at different workloads using positron emission tomography.

Skeletal muscle glucose uptake closely reflects muscle activity at exercise intensity levels <55% of maximal oxygen consumption (VO2max). Our purpose was to evaluate individual skeletal muscle activity from glucose uptake in humans during pedaling exercise at different workloads by using [18F]fluorodeoxyglucose (FDG) and positron emission tomography (PET). Twenty healthy male subjects were divided into two groups (7 exercise subjects and 13 control subjects). Exercise subjects were studied during 35 min of pedaling exercise at 40 and 55% VO2max exercise intensities. FDG was injected 10 min after the start of exercise or after 20 min of rest. PET scanning of the whole body was conducted after completion of the exercise or rest period. In exercise subjects, mean FDG uptake [standardized uptake ratio (SUR)] of the iliacus muscle and muscles of the anterior part of the thigh was significantly greater than uptake in muscles of control subjects. At 55% VO2max exercise, SURs of the iliacus muscle and thigh muscles, except for the rectus femoris, increased significantly compared with SURs at 40% VO2max exercise. Our results are the first to clarify that the iliacus muscle, as well as the muscles of the anterior thigh, is the prime muscle used during pedaling exercise. In addition, the iliacus muscle and all muscles in the thigh, except for the rectus femoris, contribute when the workload of the pedaling exercise increases from 40 to 55% VO2max.