Search

This work evaluates GPT-4V's multimodal capability for medical image analysis, focusing on three representative tasks radiology report generation, medical visual question answering, and medical visual grounding. For the evaluation, a set of prompts is designed for each task to induce the corresponding capability of GPT-4V to produce sufficiently good outputs. Three evaluation ways including quantitative analysis, human evaluation, and case study are employed to achieve an in-depth and extensive evaluation. Our evaluation shows that GPT-4V excels in understanding medical images can generate high-quality radiology reports and effectively answer questions about medical images. Meanwhile, it is found that its performance for medical visual grounding needs to be substantially improved. In addition, we observe the discrepancy between the evaluation outcome from quantitative analysis and that from human evaluation. This discrepancy suggests the limitations of conventional metrics in assessing the performance of large language models like GPT-4V and the necessity of developing new metrics for automatic quantitative analysis.

In recent years, immense developments have occurred in the field of Artificial Intelligence (AI) and the spread of broadband and ubiquitous connectivity technologies. This has led to the development and commercialization of Digital Twin (DT) technology. The widespread adoption of DT has resulted in a new network paradigm called Digital Twin Networks (DTNs), which orchestrate through the networks of ubiquitous DTs and their corresponding physical assets. DTNs create virtual twins of physical objects via DT technology and realize the co-evolution between physical and virtual spaces through data processing, computing, and DT modeling. The high volume of user data and the ubiquitous communication systems in DTNs come with their own set of challenges. The most serious issue here is with respect to user data privacy and security because users of most applications are unaware of the data that they are sharing with these platforms and are naive in understanding the implications of the data breaches. Also, currently, there is not enough literature that focuses on privacy and security issues in DTN applications. In this survey, we first provide a clear idea of the components of DTNs and the common metrics used in literature to assess their performance. Next, we offer a standard network model that applies to most DTN applications to provide a better understanding of DTN’s complex and interleaved communications and the respective components. We then shed light on the common applications where DTNs have been adapted heavily and the privacy and security issues arising from the DTNs. We also provide different privacy and security countermeasures to address the previously mentioned issues in DTNs and list some state-of-the-art tools to mitigate the issues. Finally, we provide some open research issues and problems in the field of DTN privacy and security.

Bardet–Biedl syndrome is a rare autosomal recessive genetic disorder with heterogenous clinical manifestations. The present study reports the clinical features of a novel compound heterozygous genotype of the BBS2 gene in a 14-year-old girl and her 6-year-old sister who had complaints of early-onset low vision. Fundus images revealed retinitis pigmentosa-like changes, and full-field electroretinograms showed no amplitude for the rod or cone response in both patients. Interestingly, nystagmus was observed in the older sister. On physical examination, the sisters had moderate obesity without polydactyly, hypogonadism, or intellectual disability. Exome sequencing revealed a novel compound heterozygous genotype of BBS2 in the sisters, namely the paternally inherited NM_031885.5:c.534 + 1G > T variant and the maternally inherited NM_031885.5:c.700C > T (p.Arg234Ter) variant. Both variants were classified as pathogenic according to the American College of Medical Genetics and Genomics guidelines. This study provides useful information on the genotype-phenotype relationships of the BBS2 gene for genetic counseling and diagnosis.

As one of the most promising machine learning frameworks emerging in recent years, Federated learning (FL) has received lots of attention. The main idea of centralized FL is to train a global model by aggregating local model parameters and maintain the private data of users locally. However, recent studies have shown that traditional centralized federated learning is vulnerable to various attacks, such as gradient attacks, where a malicious server collects local model gradients and uses them to recover the private data stored on the client. In this paper, we propose a decentralized federated learning against aTtacks (DEFEAT) framework and use it to defend the gradient attack. The decentralized structure adopted by this paper uses a peer-to-peer network to transmit, aggregate, and update local models. In DEFEAT, the participating clients only need to communicate with their single-hop neighbors to learn the global model, in which the model accuracy and communication cost during the training process of DEFEAT are well balanced. Through a series of experiments and detailed case studies on real datasets, we evaluate the excellent model performance of DEFEAT and the privacy preservation capability against gradient attacks.

Background: Geleophysic dysplasia and Weill-Marchesani syndrome from the acromelic dysplasias group of genetic skeletal disorders share remarkable clinical and genetic overlap.Methods: Ophthalmological, physical, radiological examinations were conducted with a female patient in her early 30 s. Whole exome sequencing followed by Sanger sequencing validation was performed to identify the genetic cause.Results: The patient, born to consanguineous Chinese parents, presented with microspherophakia, lens subluxation, high myopia, short statue, small hands and feet, stiff joints, and thickened skin. A diagnosis of Weill-Marchesani syndrome was initially made for her. However, genetic testing reveals that the patient is homozygous for the c.1966G>A (p.Gly656Ser) variant in ADAMTSL2, and that the patient’s healthy mother and daughter are heterozygous for the variant. As mutations in ADAMTSL2 are known to cause autosomal recessive geleophysic dysplasia, the patient is re-diagnosed with geleophysic dysplasia in terms of her genotype and phenotype.Conclusion: The present study describes the clinical phenotype of the homozygous ADAMTSL2 p. Gly656Ser variant, which increases our understanding of the genotype-phenotype correlation in acromelic dysplasias.

Abstract Many differential gene expression analyses are conducted with an inadequate number of biological replicates. We describe an easy and effective RNA-seq approach using molecular barcoding to enable profiling of a large number of replicates simultaneously. This approach significantly improves the performance of differential gene expression analysis. Using this approach in medaka (Oryzias latipes), we discover novel genes with sexually dimorphic expression and genes necessary for germ cell development. Our results also demonstrate why the common practice of using only three replicates in differential gene expression analysis should be abandoned.

Mobile Social Networks (MSNs) are a category of social networks that have features of both traditional OSNs and Location-Based Social Networks (LBSNs). MSNs provide users with an opportunity to explore social media, connect with people nearby, or with those sharing similar interests. The ease with which people can connect and use these networks to obtain information, share user-centric data, and request services on the go through their mobile devices as long as they have access to the internet has made MSNs omnipresent. Due to the widespread nature of MSNs and the lack of effective privacy-preserving architecture makes them a target for several attacks by adversaries and the users of the networks are completely vulnerable to these privacy breaches. With the advancements in technology, the number of attacks and their possible solutions is complex and expansive. Therefore, there is a need to properly categorize them to understand the intricacies of different attacks and the concepts used by researchers to devise different solutions. This survey on privacy issues in MSNs serves as a way to provide an introduction to MSNs. We categorize privacy in MSN on the basis of the aspect they effect and summarize the different threats in each category. Furthermore, we provide an elaborate classification of the various privacy-preserving solutions proposed in recent research works, under each category of privacy. We also review different datasets and data generation tools that can be used by future researchers in this field.

Abstract Internet of Things (IoT) is rapidly developed and widely deployed in recent years, which makes the sensory data generated by IoT systems explode. The huge amount of sensory data generated by some IoT systems has already exceeded the storage, transmission, and computation capacities of IoT systems. However, the valuable sensory data which is highly related to a query in an IoT system is relatively small. The sensory data which is highly related to a query Q forms the relative kernel dataset of Q. Therefore, retrieving sensory data in the relative kernel dataset of a query instead of the raw sensory data will reduce the heavy burdens of an IoT system in terms of transmission and computation and then reduce the energy consumption of the IoT system. In this paper, we investigate the problem of retrieving relative kernel dataset from big sensory data for continuous queries in IoT systems. Two algorithms, relative kernel dataset retrieving algorithm and piecewise linear fitting-based relative kernel dataset retrieving algorithm, are proposed to retrieve the relative kernel dataset for continuous queries. Beside, algorithms for estimating the answers of continuous queries based on their relative kernel datasets are also proposed. Extensive simulation results are provided to verify the effectiveness and energy efficiency of the proposed algorithms.

Abstract Background This investigation aimed to evaluate the role of methylation in the regulation of microRNA (miR)-122, miR-125b and miR-106b gene expression and the expression of their target genes during isoniazid (INH)-induced liver injury. Methods Rats were given INH 50 mg kg− 1·d− 1 once per day for 3, 7, 10, 14, 21 and 28 days and were sacrificed. Samples of blood and liver were obtained. Results We analysed the methylation and expression levels of miR-122, miR-125b and miR-106b and their potential gene targets in livers. Liver tissue pathologies, histological scores and alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities changed, indicating the occurrence of liver injury. Relative expression levels of miR-122, miR-125b and miR-106b genes in the liver decreased after INH administration and correlated with the scores of liver pathology and serum AST and ALT activities, suggesting that miR-122, miR-125b and miR-106b are associated with INH-induced liver injury. The amount of methylated miR-122, miR-125b and miR-106b in the liver increased after INH administration and correlated with their expression levels, suggesting the role of methylation in regulating miRNA gene expression. Two miR-122 gene targets, cell cycle protein G1 (Cyclin G1) and cationic amino acid transporter-1 (CAT-1), also increased at the mRNA and protein levels, which suggests that lower levels of miR-122 contribute to the upregulation of Cyclin G1 and CAT-1 and might play a role in INH-induced liver injury. Signal transducer and activator of transcription 3 (STAT3) was a common target gene of miR-125b and miR-106b, and its expression levels of mRNA and protein increased after INH administration. The protein expression of phosphorylated (p)-STAT3 and the mRNA expression of RAR-related orphan receptor gamma (RORγt) regulated by p-STAT3 also increased. Meanwhile, the mRNA and protein expression of interleukin (IL)-17 regulated by RORγt, and the mRNA and protein expression of CXCL1 and MIP-2 regulated by IL-17 increased after INH administration. These results demonstrate that lower levels of hepatic miR-125b and miR-106b contribute to the upregulation of STAT3 in stimulating the secretion of inflammatory factors during INH-induced liver injury. Conclusions Our results suggested that DNA methylation probably regulates the expression of miRNA genes (miR-122, miR-125b, and miR-106b), affecting the expression of their gene targets (Cyclin G1, CAT-1, and STAT3) and participating in the process of INH-induced liver injury.

With the rapid development in mobile devices, mobile crowdsourcing has become an important research focus. In large-scale mobile crowdsourcing, the effective evolution prediction and incentive mechanism are the key focuses to improve the efficiency of systems. The evolution model based on evolutionary game theory is researched to predict the evolution trends of mobile crowdsourcing systems effectively. Based on the evolution trends, the reputation updating mechanism (RUMG) is proposed to address free-riding and false-reporting problems. According to spatio-temporal privacy preserving, the incentive mechanism with spatio-temporal privacy preserving for mobile crowdsourcing is researched. In order to protect worker's spatio-temporal privacy information effectively, a spatio-temporal privacy preserving based on k-anonymity (LKAC) is proposed. In addition, the effectivenesses of the proposed RUMG and LKAC are verified through comparison experiments. This proposed mechanism also improves the security of system and resolves the free-riding and false-reporting problems of mobile crowdsourcing.

Privacy in social networks is a large and growing concern in recent times. It refers to various issues in a social network which include privacy of users, links, and their attributes. Each privacy component of a social network is vast and consists of various sub-problems. For example, user privacy includes multiple sub-problems like user location privacy, and user personal information privacy. This survey on privacy in social networks is intended to serve as an initial introduction and starting step to all further researchers. We present various privacy preserving models and methods include naive anonymization, perturbation, or building a complete alternative network. We show the work done by multiple researchers in the past, where social networks are stated as network graphs with users represented as nodes and friendship between users represented as links between the nodes. We study ways and mechanisms developed to protect these nodes and links in the network. We also review other systems proposed, along with all the available databases for future researchers in this area.

Abstract This paper investigates the problem of maximizing utility for job scheduling where each job consists of multiple tasks, each task has utility and each job also has extra utility if all tasks of that job are completed. We provide a 2-approximation algorithm for the single-machine case and a 2-approximation algorithm for the multi-machine problem. Both algorithms include two steps. The first step employs the Earliest Deadline First method to compute utility with only extra job utility, and it is proved that it obtains the optimal result for this sub-problem. The second step employs a Dynamic Programming method to compute utility without extra job utility, and it also derives the optimal result. An approximation result can then be obtained by combining the results of the two steps.

Location-based services (LBS) leveraged by ubiquitous mobile devices have brought great convenience to mobile users in various aspects, including communication, information exchange, social activities, and so on. However, privacy concerns arise at the same time, since the users need to submit their locations and query contents to the LBS servers. To this end, location privacy and query privacy have been recognized. In particular, this paper focuses on query privacy for preventing the leakage of users' query contents. Cloaking region-based techniques using untrusted servers and client-based k-anonymity approaches have been devised to preserve query privacy in LBS. However, these works suffer from single point of failure or insufficiency of query privacy. To address this issue, we investigate effective k-anonymity-based solutions for query privacy in LBS. We formulate a probabilistic framework PkA, under which k-anonymity-based mechanisms can be initiated, and analyze a recent proposed algorithm DLS as an instance of PkA. An algorithm circle segment is presented to provide effective query privacy when query interests have similar prior probability. To obtain more effective query privacy in general cases, we propose two algorithms MEE and MER, which optimize two individual privacy metrics, denoted expected entropy and expected max-min ratio, adopted in this paper. We recognize two practical properties - No More Leakage and k-Effectiveness for effective query privacy, and our proposed algorithms satisfy both of No More Leakage and k-Effectiveness. We conduct evaluation based on real-life data sets and synthetic distributions of query interests, and the evaluation results demonstrate that our proposed algorithms produce significantly improved query privacy.

The inherent stochasticity generates substantial gene expression variation among isogenic cells under identical conditions, which is frequently referred to as gene expression noise or cell-to-cell expression variability. Similar to (average) expression level, expression noise is also subject to natural selection. Yet it has been observed that noise is negatively correlated with expression level, which manifests as a potential constraint for simultaneous optimization of both. Here, we studied expression noise in human embryonic cells with computational analysis on single-cell RNA-seq data and in yeast with flow cytometry experiments. We showed that this coupling is overcome, to a certain degree, by a histone modification strategy in multiple embryonic developmental stages in human, as well as in yeast. Importantly, this epigenetic strategy could fit into a burst-like gene expression model: promoter-localized histone modifications (such as H3K4 methylation) are associated with both burst size and burst frequency, which together influence expression level, while gene-body-localized ones (such as H3K79 methylation) are more associated with burst frequency, which influences both expression level and noise. We further knocked out the only "writer" of H3K79 methylation in yeast, and observed that expression noise is indeed increased. Consistently, dosage sensitive genes, such as genes in the Wnt signaling pathway, tend to be marked with gene-body-localized histone modifications, while stress responding genes, such as genes regulating autophagy, tend to be marked with promoter-localized ones. Our findings elucidate that the "division of labor" among histone modifications facilitates the independent regulation of expression level and noise, extend the "histone code" hypothesis to include expression noise, and shed light on the optimization of transcriptome in evolution.

With the rapid development of mobile devices, mobile crowdsourcing has become an important research focus. According to the task allocation, scholars have proposed many methods. However, few works discuss combining social networks and mobile crowdsourcing. To maximize the utilities of mobile crowdsourcing system, this paper proposes a task allocation model considering the attributes of social networks for mobile crowdsourcing system. Starting from the homogeneity of human beings, the relationship between friends in social networks is applied to mobile crowdsourcing system. A task allocation algorithm based on the friend relationships is proposed. The GeoHash coding mechanism is adopted in the process of calculating the strength of worker relationship, which effectively protects the location privacy of workers. Utilizing synthetic dataset and the real-world Yelp dataset, the performance of the proposed task allocation model was evaluated. Through comparison experiments, the effectiveness and applicability of the proposed allocation mechanism were verified.

In order to avoid malicious competition and select high quality crowd workers to improve the utility of crowdsourcing system, this paper proposes an incentive mechanism based on the combination of reverse auction and multi-attribute auction in mobile crowdsourcing. The proposed online incentive mechanism includes two algorithms. One is the crowd worker selection algorithm based on multi-attribute reverse auction that adopts dynamic threshold to make an online decision for whether accept a crowd worker according to its attributes. Another is the payment determination algorithm which determines payment for a crowd worker based on its reputation and quality of sensing data, that is, a crowd worker can get payment equal to the bidding price before performing task only if his reputation reaches good reputation threshold, otherwise he will get payment based on his data sensing quality. We prove that our proposed online incentive mechanism has the properties of computational efficiency, individual rationality, budget-balance, truthfulness and honesty. Through simulations, the efficiency of our proposed online incentive mechanism is verified which can improve the efficiency, adaptability and trust degree of the mobile crowdsourcing system.

The retransmission threshold in wireless sensor networks is critical to the latency of data delivery in the networks. However, existing works on data transmission in sensor networks did not consider the optimization of the retransmission threshold, and they simply set the same retransmission threshold for all sensor nodes in advance. The method did not take link quality and delay requirement into account, which decreases the probability of a packet passing its delivery path within a given deadline. This paper investigates the problem of finding optimal retransmission thresholds for relay nodes along a delivery path in a sensor network. The object of optimizing retransmission thresholds is to maximize the summation of the probability of the packet being successfully delivered to the next relay node or destination node in time. A dynamic programming-based distributed algorithm for finding optimal retransmission thresholds for relay nodes along a delivery path in the sensor network is proposed. The time complexity is O n Δ · max 1 ≤ i ≤ n { u i } , where u i is the given upper bound of the retransmission threshold of sensor node i in a given delivery path, n is the length of the delivery path and Δ is the given upper bound of the transmission delay of the delivery path. If Δ is greater than the polynomial, to reduce the time complexity, a linear programming-based ( 1 + p m i n ) -approximation algorithm is proposed. Furthermore, when the ranges of the upper and lower bounds of retransmission thresholds are big enough, a Lagrange multiplier-based distributed O ( 1 ) -approximation algorithm with time complexity O ( 1 ) is proposed. Experimental results show that the proposed algorithms have better performance.

With the emergence of new technologies, mobile devices are capable of undertaking computational and sensing tasks. A large number of users with these mobile devices promote the formation of the Mobile Crowdsourcing Systems (MCSs). Within a MCS, each mobile device can contribute to the crowdsourcing platform and get rewards from it. In order to achieve better performance, it is important to design a mechanism that can attract enough participants with mobile devices and then allocate the tasks among participants efficiently. In this paper, we are interested in the investigation of tasks allocation and price determination in MCSs. Two truthful auction mechanisms are proposed for different working patterns. A Vickrey–Clarke–Groves (VCG)-based auction mechanism is proposed to the continuous working pattern, and a suboptimal auction mechanism is introduced for the discontinuous working pattern. Further analysis shows that the proposed mechanisms have the properties of individual rationality and computational efficiencies. Experimental results suggest that both mechanisms guarantee all the mobile users bidding with their truthful values and the optimal maximal social cost can be achieved in the VCG-based auction mechanism.