Background: Classification of infants has always been considered a crucial task in the literature related to predicting small for gestational age (SGA) infants. Traditional medical guidance becomes increasingly unsatisfactory, as patients' care should be centered not only on clinical symptoms but also on socio-economic and demographic factors. Infants with excessive gestational weight exhibit serious maternal complications that require early intervention to stream-line the progression of the disease. Methods: This research proposes to use the Stacked Generalization Scheme (SGS) to predict Small for Gestational (SGA) Infants on the dataset collected from the National Pre-Pregnancy and Examination Program of China. A Cleaned Feature Vector (CFV) is created that entertains issues related to missing values, discretization of fields, and data imbalance. Later, Knowledge-Driven Data (KDD) Features are extracted from the obtained CFV, and the proposed scheme is applied to predict SGA infants. The proposed scheme superposed the existing baseline approaches by achieving the highest precision, recall, and AUC scores of 0.94, 0.85, and 0.89, respectively. Conclusion: The proposed SGS can predict SGA infants accurately compared to existing baseline schemes using KDD parameters, which can help pediatricians develop an efficient SGA Prognosis process.

Myopia has become a major public health problem affecting the eye health of our citizens, especially teenagers. Fundus images can be obtained non-invasively and can be used to monitor and follow up on the progress in high myopia. However, with the development of artificial intelligence, it is still difficult to establish a computer-aided diagnosis model for high myopia with young children as research objects, mainly because 1) it is very difficult to collect labeled fundus images, and there is no large amount of such data that can be freely accessed; 2) hard samples which have clinical significance for population screening and diagnosis are rare and indistinguishable. To solve these problems, we propose a high myopia detection model on fundus images based on curriculum learning. We design a dual-curriculum generation module, which aims to use the expert model to endow the curriculum with new training indicators so that the student model can gradually and robustly identify hard samples. Compared with the baseline, our framework significantly improves the convergence speed of the training process and achieves the best performance during testing. Experimental results on a high myopia fundus images dataset show that our framework provides efficient and accurate detection and outperforms other methods.

Recently, lexicon-based Chinese Named Entity Recognition (NER) models have achieved state-of-the-art performance by benefiting from the rich boundary and semantic information contained in the lexicon. However, in the Chinese medical domain, it’s difficult to obtain the medical lexicon related to the target medical corpus. In this paper, we propose a new paradigm, enhancing Chinese medical NER with Auto-mined Lexicon (ALNER), which alleviates the difficulty of obtaining the medical lexicon by designing a data-driven automatic lexicon construction method. We define medical lexicon construction as a high-quality phrase mining task. We perform secondary annotation on the NER annotated data and use the secondary annotated data to train a deep learning-based phrase tagger. Experimental results show that our method can be combined with different lexicon-based Chinese NER models to improve performance and that the method does not require an external medical lexicon.

The wide adoption of electronic medical record (EMR) systems causes rapid growth of medical and clinical data. It makes the medical named entity recognition (NER) technologies become critical to find useful patient information in the medical dataset. However, the medical terminologies usually have the characteristics of inherent complexity and ambiguity, it is difficult to capture context-dependency representations by supervision signal from a simple single layer structure model. In order to address this problem, this paper proposes a hybrid model based on stacked Bidirectional Long Short-Term Memory (BILSTM) for medical named entity recognition, which we call BSBC (BERT combined with stacked BILSTM and CRF). First, we use Bidirectional Encoder Representation from Transformers (BERT) to perform unsupervised learning on an unlabeled dataset to obtain character-level embeddings. Then, stacked BILSTM is utilized to obtain context-dependency representations through the multi hidden layers structure. Finally, Conditional Random Field (CRF) is used to predict sequence tags. The experiment results show that our method significantly outperforms the baseline methods, it serves as a strong alternative approach compared with traditional methods.

It has been estimated that the global gaming market is worth nearly US$150 billion. Its consumer chargeback services often end up being used by some online gamers as a tool to commit fraud, causing a huge adverse impact on the industry. A gaming company in Taiwan found itself falling victim of malicious chargeback fraud. Nearly NT$10 million of fraudulent chargebacks were made during the period from January to April 2019 alone, making a huge dent in the revenue of the company. To counter chargeback fraud, some gaming companies resorted to manually checking for and blocking malicious accounts of their users, incurring huge labor cost in the process. Manual checking might have alleviated the problems to some extent; however, when new games came online, gaming companies would see a surge of malicious chargebacks, causing subsequent exponential increases in losses. To help reduce labor cost incurred by manual account checking, potential human errors and potential losses that may be caused by malicious chargebacks, this study proposed a k-NN model to detect malicious chargebacks by analysing online gamers' transactional records and gameplay data. The numbers of times and the amounts of prepayment, the numbers of times of chargebacks, and the times of the transactions that the gamers of our study gaming company made were used as characteristics for our k-NN model. The use of these characteristics enabled us to score a minimum of 0.81 in F1-Measure. In addition, three SMOTE (Synthetic Minority Over-sampling Technique) sampling methods were used to deal with the imbalance data provided by our study company and improve the F1-Measure of our proposed k-NN model (scoring up to 0.89 in our experiments). It is hoped that the use of our k-NN model can help reduce potential losses of online gaming companies that may be caused by malicious chargeback fraud, deter to malicious gamers against illegal gains, and prevent the online gaming ecosystem from being sabotaged by malicious chargebacks.

In existing trust establishment researches, the decision making accuracy and decision delay of received event message are both the outstanding problems in VANETs. Especially on the decision delay, the delay of alert event might lead to traffic accident and injury. In this paper, we propose an adaptive decision making model which can make decision quickly by the effective and quick assistance of the RSUs. To evaluate the detection accuracy and efficiency of our proposed system, we conducted several simulations under different trust attacks. In the experiment results, the purpose model can shorten decision delay while raising the detection accuracy.

In VANETs, how to determine the trustworthiness of event messages has received a great deal of attentions in recent years for improving the safety and location privacy of vehicles. Among these research studies, the accuracy and delay of trustworthiness decision are both important problems. In this paper, we propose a road-side unit (RSU) and beacon-based trust management system, called RaBTM, which aims to prorogate message opinions quickly and thwart internal attackers from sending or forwarding forged messages in privacy-enhanced VANETs. To evaluate the performance and efficiency of the proposed system, we conducted a set of simulations under alteration attacks and bogus message attacks with various adversary ratios. The simulation results show that the proposed system RaBTM is highly resilient to adversarial attacks and performs at least 15% better than weighted vote (WV) scheme.

In recent years more and more studies are focused on trust management of vehicle ad-hoc networks (VANETs). However, little attention has been given to the issue of location privacy of the existing trust methodologies in the literature. Although traffic safety remains to be the most crucial issue in VANETs, location privacy can be just as important for drivers, and neither of them can be ignored. In this paper, we propose a trust scheme which aims to thwart internal attackers in privacy enhanced VANETS. In the proposed scheme a secure broadcast authentication protocol and beacon-based trust management system are being employed to maintain the trustworthiness of vehicles. We adopt Dempster-Shafer Theory to incorporate trustworthiness of event message with vehicle trustworthiness from multiple vehicles. In order to ensure the reliability of the proposed scheme, we evaluate the performance under alteration and denial-of-service attack models. The simulation results show that the proposed system is highly resilient to adversary attacks no matter whether it is under fixed silent period (FSP) scheme or random silent period (RSP) location privacy enhancement scheme.

To enhance driving safety in vehicle ad-hoc networks (VANETs), vehicles periodically broadcast safety messages with information of their precise positions to others. These broadcast messages, however, make it easy to track vehicles and will likely lead to violations of personal privacy. Unfortunately, most of the current location privacy enhancement methodologies in VANETs suffer some shortcomings and do not take driving safety into consideration. In this paper, we propose a safe distance based location privacy scheme called SafeAnon, which can significantly enhance location privacy as well as traffic safety.