Search

One way to diagnose COVID-19 is to use the Polymerase Chain Reaction (PCR) test. However, this test is rather invasive. An alternative would be to use chest images of the patients to diagnose if the patient has COVID-19. These chest X-ray images have to be manually annotated by a medical professional such as a radiologist, and due to privacy concerns, getting access to readily available and annotated COVID-19 chest X-ray images is difficult. In order to train a deep learning model to perform image classification tasks, it is prudent to train the deep-learning model on a large enough dataset to avoid the problem of overfitting. In this paper, we explore using Generative Adversarial Networks (GANs) as a form of data augmentation technique to enlarge the training data for deep learning models. We first explored how the synthetic data generated by GANs are affected by the training size. Following which, we compared the performance of the two different GAN architectures, namely the Deep Convolutional Generative Adversarial Networks (DCGAN) and the Wasserstein Generative Adversarial Networks with Gradient Penalty (WGAN-GP). We successfully used GANs to generate synthetic COVID-19 chest X-ray images with a Fréchet Inception Distance (FID) score that was below 2. [ABSTRACT FROM AUTHOR]

We develop a method for analyzing multivariate time series using topological data analysis (TDA) methods. The proposed methodology involves converting the multivariate time series to point cloud data, calculating Wasserstein distances between the persistence diagrams and using the k -nearest neighbours algorithm ( k -NN) for supervised machine learning. Two methods (symmetry-breaking and anchor points) are also introduced to enable TDA to better analyze data with heterogeneous features that are sensitive to translation, rotation or choice of coordinates. We apply our methods to room occupancy detection based on 5 time-dependent variables (temperature, humidity, light, CO2 and humidity ratio). Experimental results show that topological methods are effective in predicting room occupancy during a time window. We also apply our methods to an Activity Recognition dataset and obtained good results. [ABSTRACT FROM AUTHOR]

Introduction Labour induction in women with a previous caesarean delivery currently uses vaginal prostaglandin E2 (PGE2), which carries the risks of uterine hyperstimulation and scar rupture. We aim to compare the efficacy of mechanical labour induction using a transcervically applied Foley catheter balloon (FCB) with PGE2 in affected women attempting trial of labour after caesarean (TOLAC). Methods and analysis This single-centre non-inferiority prospective, randomised, open, blinded-endpoint study conducted at an academic maternity unit in Singapore will recruit a total of 100 women with one previous uncomplicated caesarean section and no contraindications to vaginal delivery. Eligible consented participants with term singleton pregnancies and unfavourable cervical scores (≤5) requiring labour induction undergo stratified randomisation based on parity and are assigned either FCB (n=50) or PGE2 (n=50). Treatments are applied for up to 12 hours with serial monitoring of the mother and the fetus and serial assessment for improved cervical scores. If the cervix is still unfavourable, participants are allowed a further 12 hours’ observation for cervical ripening. Active labour is initiated by amniotomy at cervical scores of ≥6. The primary outcome is the rate of change in the cervical score, and secondary outcomes include active labour within 24 hours of induction, vaginal delivery, time-to- delivery interval and uterine hyperstimulation. All analyses will be intention-to- treat. The data generated in this trial may guide a change in practice towards mechanical labour induction if this proves efficient and safer for women attempting TOLAC compared with PGE2, to improve labour management in this high-risk population. Ethics and dissemination Ethical approval is granted by the Domain Specific Review Board (Domain D) of the National Healthcare Group, Singapore. All adverse events will be reported within 24 hours of notification for assessment of causality. Data will be published and will be available for future meta-analyses. [ABSTRACT FROM AUTHOR]

This article first reviews business analytics (BA) process with its key tasks, and then proposes a framework of an evolving multiagent system from the perspective of an automated business analytics. Under this view, typical BA tasks, such as prediction, evaluation, and decision are carried out by individual intelligent agents in a multiagent system (MAS) with an evolving environment to cope with the target business application. [ABSTRACT FROM AUTHOR]