Bone mineral density in haemophilia patients: A systematic review and meta-analysis.

INTRODUCTION: With the increase in life expectancy of haemophilia patients (PWH), the risk of osteoporosis increases, but there is little research on whether haemophilia is the cause of osteoporosis. AIM: To conduct systematically review whether bone mineral density (BMD) in PWH decreased and the factors affecting BMD. METHODS: Two authors independently searched databases and reviewed citations from relevant articles, selecting studies published in any language and performed in humans before March 2023. Eligibility criteria were observational studies in PWH, with BMD as at least one outcome other than osteoporosis or bone loss, and analyses in a group of PWH and healthy controls. RESULTS: Twelve studies were ultimately identified, consisting of 1210 individuals (534 PWH and 676 healthy controls), compared with the control group, BMD in PWH decreased by 0.13 g/cm2 [95% confidence interval (CI) -0.18 to -0.08, I2  = 89%]. No evidence of publication bias was detected. There was no evidence that age, BMI, level of physical activity, the types of haemophilia, haemophilia severity, a blood-borne virus (HCV) and treatment modality predicted the BMD in PWH. CONCLUSION: The results indicate that BMD in PWH is lower than in healthy controls. Therefore, we strongly recommend PWH early measurement of BMD to prevent osteoporosis.

A Systematic Review and Meta-Analysis of Influences of Chronic Kidney Disease on Patients after Percutaneous Coronary Intervention for Chronic Total Occlusions.

Objective: Chronic kidney disease (CKD) is a clinical collective term for kidney disease with glomerular filtration rate (GFR) < 60 mL/min for more than three months due to various factors and is usually associated with coronary heart disease and is also an independent risk factor for coronary heart disease. This study is aimed at systematically reviewing the influence of CKD on the outcomes of patients after percutaneous coronary intervention (PCI) for chronic total occlusions (CTOs). Methods: The Cochrane Library, PubMed, Embase, China biomedical literature database (SinoMed), China National Knowledge Infrastructure, and Wanfang database were searched for case-control studies on the influence of CKD on outcomes after PCI for CTOs. After screening the literature, extracting data, and evaluating the quality of literature, RevMan 5.3 software was used for meta-analysis. Results: There were 11 articles with a total of 558,440 patients included. Meta-analysis results indicated that left ventricular ejection fraction (LVEF) level, diabetes, smoking, hypertension, coronary artery bypass grafting, angiotensin converting enzyme inhibitor (ACEI)/angiotensin receptor blocker (ARB), ß-blockers, age, and renal insufficiency were the factors affecting outcomes after PCI for CTOs [risk ratio and 95% confidence interval were: 0.88 (0.86, 0.90), 0.96 (0.95, 0.96), 0.76 (0.59, 0.98), 1.39 (0.89, 2.16), 0.73 (0.38, 1.40), 0.24 (0.02, 3.9), 0.78 (0.77, 0.79), 0.81 (0.80, 0.82), and 1.50 (0.47, 4.79)]. Conclusion: LVEF level, diabetes, smoking, hypertension, coronary artery bypass grafting, ACEI/ARB, ß-blockers, age, renal insufficiency, etc. are important risk factors for outcomes after PCI for CTOs. Controlling these risk factors is of great significance for the prevention, treatment, and prognosis of CKD.

Contributing Factors of Dielectric Properties for Polymer Matrix Composites.

Due to the trend of multi-function, integration, and miniaturization of electronics, traditional dielectric materials are difficult to satisfy new requirements, such as balanced dielectric properties and good designability. Therefore, high dielectric polymer composites have attracted wide attention due to their outstanding processibility, good designability, and dielectric properties. A number of polymer composites are employed in capacitors and sensors. All these applications are directly affected by the composite's dielectric properties, which are highly depended on the compositions and internal structure design, including the polymer matrix, fillers, structural design, etc. In this review, the influences of matrix, fillers, and filler arrangement on dielectric properties are systematically and comprehensively summarized and the regulation strategies of dielectric loss are introduced as well. Finally, the challenges and prospects of high dielectric polymer composites are proposed.

Bach1 modulates AKT3 transcription to participate in hyperglycaemia-mediated EndMT in vascular endothelial cells.

Hyperglycaemia-mediated endothelial-to-mesenchymal transition (EndMT) is involved in the occurrence and progression of cardiovascular complications in diabetic patients. Previous studies reported that AKT serine/threonine kinase 3 (AKT3) and Bric-a-brac/Tramtrack/Broad (BTB) and cap'n'collar (CNC) homology 1 (bach1) participates in endothelial injury and epithelial-to-mesenchymal transition. In the present study, we proposed that bach1 regulates AKT3 transcription, thus involved in hyperglycaemia-mediated EndMT in vascular endothelium. Our results indicated that hyperglycaemia/high glucose increased AKT3 expression and induced EndMT in aorta of diabetic rats and hyperglycaemic human umbilical vein endothelial cells (HUVECs). Moreover, inhibition of AKT3 expression reversed high glucose-mediated EndMT in HUVECs. Further, hyperglycaemia/high glucose augmented bach1 expression in aorta of diabetic rats and hyperglycaemic HUVECs. Furthermore, si-bach1 countered high glucose-induced AKT3 expression and EndMT in HUVECs. In addition, the effect of bach1 overexpression is similar to that of high glucose treatment, which was reversed by si-AKT3. ChIP assays found bach1 enriched in the promoter region of AKT3. Bach1 overexpression augmented AKT3 promoter activity, which lost after specific binding site mutation. Bach1 was involved in hyperglycaemia-induced EndMT via modulation of AKT3 transcription.

Differentially Expressed microRNAs in Peritoneal Dialysis Effluent-Derived Exosomes from the Patients with Ultrafiltration Failure.

Background: Ultrafiltration failure remains one of the most severe complications of long-term peritoneal dialysis (PD), which results in death. This study aimed to characterize the circulating exosomal microRNA (miRNA) profiles associated with ultrafiltration failure and explore its underlying mechanisms. Methods: Exosomes were isolated from the peritoneal dialysis effluent (PDE) of patients with ultrafiltration failure or success using the ultracentrifugation method, and then transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and western blot were used for exosome characterization. After that, the isolated exosomes were sent for small RNA sequencing, and eight differentially expressed miRNAs (DE-miRNAs) were chosen for RT-qPCR validation. Results: TEM, NTA, and western blot revealed that exosomes were successfully isolated. After sequencing, 70 DE-miRNAs involved in ultrafiltration were identified, including 41 upregulated ones and 29 downregulated ones. Functional analyses revealed that these DE-miRNAs were significantly enriched in pathways of cancer, ubiquitin-mediated proteolysis, axon orientation, and the Rap1 and Ras signaling pathways. In addition, the consistency rate of RT-qPCR and sequencing results was 75%, which indicated the relatively high reliability of the sequencing data. Conclusions: Our findings implied that these DE-miRNAs may be potential biomarkers of ultrafiltration failure, which would help us to discover novel therapeutic targets/pathways for ultrafiltration failure in patients with end-stage renal disease.

Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction.

Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors.

Multi-Source Selection Transfer Learning with Privacy-Preserving.

Transfer learning has ability to create learning task of weakly labeled or unlabeled target domain by using knowledge of source domain to help, which can effectively improve the performance of target learning task. At present, the increased awareness of privacy protection restricts access to data sources and poses new challenges to the development of transfer learning. However, the research on privacy protection in transfer learning is very rare. The existing work mainly uses differential privacy technology and does not consider the distribution difference between data sources, or does not consider the conditional probability distribution of data, which causes negative transfer to harm the effect of algorithm. Therefore, this paper proposes multi-source selection transfer learning algorithm with privacy-preserving MultiSTLP, which is used in scenarios where target domain contains unlabeled data sets with only a small amount of group probability information and multiple source domains with a large number of labeled data sets. Group probability means that the class label of each sample in target data set is unknown, but the probability of each class in a given data group is available, and multiple source domains indicate that there are more than two source domains. The number of data set contains more than two data sets of source domain and one data set of target domain. The algorithm adapts to the marginal probability distribution and conditional probability distribution differences between domains, and can protect the privacy of target data and improve classification accuracy by fusing the idea of multi-source transfer learning and group probability into support vector machine. At the same time, it can select the representative dataset in source domains to improve efficiency relied on speeding up the training process of algorithm. Experimental results on several real datasets show the effectiveness of MultiSTLP, and it also has some advantages compared with the state-of-the-art transfer learning algorithm.

The efficacy of modified HuangLian JieDu decoction for early enteral nutrition in patients with sepsis: A randomized controlled study.

OBJECTIVE: This study aimed to evaluate the efficacy of modified HuangLian JieDu decoction (MHLJDD) as a supplementary medication for early enteral nutrition in septic patients. METHODS: This study was designed as a randomized controlled preliminary study. Septic patients were randomly divided into control (treated with the base treatment) and intervention (co-treated with MHLJDD and the base treatment) groups. The primary outcomes of this study were 60-day (d) mortality rate, length of mechanical ventilation (MV), and length of stay in the intensive care unit (ICU). RESULTS: Of the 86 included patients, 44 and 42 were allocated to the intervention and control groups, respectively. Lengths of MV and ICU stay were significantly shorter in the intervention group than in the control group (10.31 ±â€…3.92 d vs 8.66 ±â€…2.84 d, P = .028; and 11.88 ±â€…5.25 d vs 10.41 ±â€…3.14 d, P = .029; respectively). However, the difference in 60-d mortality rate between the 2 groups was not statistically significant (20.45% vs 38.10%, P = .071). The enteral-nutrition tolerance score of the control group was higher than that of the intervention group (6.81 ±â€…4.28 vs 4.68 ±â€…4.04, P = .020). Incidence of hyperglycemia and gastric retention (gastric residual volume > 250 mL) was higher in the control group than in the intervention group (59.52% vs 29.55%, P = .005; and 28.57% vs 11.36%, P = .020, respectively). CONCLUSIONS: MHLJDD can shorten the MV and ICU stay of septic patients.

Multi-source fast transfer learning algorithm based on support vector machine.

Knowledge in the source domain can be used in transfer learning to help train and classification tasks within the target domain with fewer available data sets. Therefore, given the situation where the target domain contains only a small number of available unlabeled data sets and multi-source domains contain a large number of labeled data sets, a new Multi-source Fast Transfer Learning algorithm based on support vector machine(MultiFTLSVM) is proposed in this paper. Given the idea of multi-source transfer learning, more source domain knowledge is taken to train the target domain learning task to improve classification effect. At the same time, the representative data set of the source domain is taken to speed up the algorithm training process to improve the efficiency of the algorithm. Experimental results on several real data sets show the effectiveness of MultiFTLSVM, and it also has certain advantages compared with the benchmark algorithm.

circAMOTL1L Suppresses Renal Cell Carcinoma Growth by Modulating the miR-92a-2-5p/KLLN Pathway.

Accumulating evidence indicates that the dysregulation of circular RNAs (circRNAs) contributes to tumor progression; however, the regulatory functions of circRNAs in renal cell carcinoma (RCC) remain largely unknown. In this study, the function and underlying mechanism of circAMOTL1L in RCC progression were explored. qRT-PCR showed the downregulation of circAMOTL1L in RCC tissues and cell lines. The decrease in circAMOTL1L expression correlated with the tumor stage, metastasis, and poor prognosis in patients with RCC. Functional experiments revealed that circAMOTL1L inhibited cell proliferation and increased apoptosis in RCC cells. Subcutaneous implantation with circAMOTL1L-overexpressing cells in nude mice decreased the growth ability of the xenograft tumors. Mechanistically, circAMOTL1L served as a sponge for miR-92a-2-5p in upregulating KLLN (killin, p53-regulated DNA replication inhibitor) expression validated by bioinformatics analysis, oligo pull-down, and luciferase assays. Further, reinforcing the circAMOTL1L-miR-92a-2-5p-KLLN axis greatly reduced the growth of RCC in vivo. Conclusively, our findings demonstrate that circAMOTL1L has an antioncogenic role in RCC growth by modulating the miR-92a-2-5p-KLLN pathway. Thus, targeting the novel circAMOTL1L-miR-92a-2-5p-KLLN regulatory axis might provide a therapeutic strategy for RCC.

Comparison of outcomes of peritoneal dialysis between patients after failed kidney transplant and transplant-naïve patients: a meta-analysis of observational studies.

PURPOSE: The influence of prior failed kidney transplants on outcomes of peritoneal dialysis (PD) is unclear. Thus, we conducted a systematic review and meta-analysis to compare the outcomes of patients initiating PD after a failed kidney transplant with those initiating PD without a prior history of kidney transplantation. METHODS: We searched PubMed, Embase, CENTRAL, and Google Scholar databases from inception until 25 November 2020. Our meta-analysis considered the absolute number of events of mortality, technical failures, and patients with peritonitis, and we also pooled multi-variable adjusted hazard ratios (HR). RESULTS: We included 12 retrospective studies. For absolute number of events, our analysis indicated no statistically significant difference in technique failure [RR, 1.14; 95% CI, 0.80-1.61; I2=52%; p = 0.48], number of patients with peritonitis [RR, 1.13; 95% CI, 0.97-1.32; I2=5%; p = 0.11] and mortality [RR, 1.00; 95% CI, 0.67-1.50; I2=63%; p = 0.99] between the study groups. The pooled analysis of adjusted HRs indicated no statistically significant difference in the risk of technique failure [HR, 1.25; 95% CI, 0.88-1.78; I2=79%; p = 0.22], peritonitis [HR, 1.04; 95% CI, 0.72-1.50; I2=76%; p = 0.85] and mortality [HR, 1.24; 95% CI, 0.77-2.00; I2=66%; p = 0.38] between the study groups. CONCLUSION: Patients with kidney transplant failure initiating PD do not have an increased risk of mortality, technique failure, or peritonitis as compared to transplant-naïve patients initiating PD. Further studies are needed to evaluate the impact of prior and ongoing immunosuppression on PD outcomes.

Specific microstructural changes of the cervical spinal cord in syringomyelia estimated by diffusion tensor imaging.

The microstructure of the spinal cord in syringomyelia has not been well studied. The aim of this study was to evaluate the microstructure of the cervical cord in patients with syringomyelia using diffusion tensor imaging (DTI) and to investigate the association between DTI parameters and the size of the syrinx cavity. Thirty patients with syringomyelia and 11 age-matched controls were included in this study. DTI and T1/T2-weighted MRI were used to estimate spinal microstructure. The patients were divided into a clinical symptom group (group A) and a non-clinical symptom group (group B) according to ASIA assessments. The fractional anisotropy (FA) and apparent diffusion coefficient (ADC) values (mm2/s) were measured and compared between patients and controls. Correlation between FA/ADC and the size of the syrinx cavity was examined with a bivariate analysis. FA values were lower (P < 0.000) and ADC values were higher (P < 0.000) compared to the controls at the level of all syrinxes examined in patients with syringomyelia; both FA values and ADC values reached normal values either above or below the syrinx levels (all P > 0.05). FA values and ADC values at all cervical levels were not significantly different either in controls or outside of the syrinx (all P > 0.05). FA values of group A was significantly lower than those of group B (P < 0.000). There was a negative association between FA values and the size of syrinx cavity, and a positive association between ADC values and the size of syrinx cavity (FA: P < 0.05, ADC: P < 0.05). The microstructure of the cervical spinal cord is different across all patients with syringomyelia. DTI is a promising tool for estimating quantitative pathological characteristics that are not visible with general MRI.

Tensor Gradient L0-Norm Minimization-Based Low-Dose CT and Its Application to COVID-19.

Methods to recover high-quality computed tomography (CT) images in low-dose cases will be of great benefit. To reach this goal, sparse-data subsampling is one of the common strategies to reduce radiation dose, which is attracting interest among the researchers in the CT community. Since analytic image reconstruction algorithms may lead to severe image artifacts, the iterative algorithms have been developed for reconstructing images from sparsely sampled projection data. In this study, we first develop a tensor gradient L0-norm minimization (TGLM) for low-dose CT imaging. Then, the TGLM model is optimized by using the split-Bregman method. The Coronavirus Disease 2019 (COVID-19) has been sweeping the globe, and CT imaging has been deployed for detection and assessing the severity of the disease. Finally, we first apply our proposed TGLM method for COVID-19 to achieve low-dose scan by incorporating the 3-D spatial information. Two COVID-19 patients (64 years old female and 56 years old man) were scanned by the [Formula: see text]CT 528 system, and the acquired projections were retrieved to validate and evaluate the performance of the TGLM.

Microstructural Changes in Compressed Cervical Spinal Cord Are Consistent With Clinical Symptoms and Symptom Duration.

STUDY DESIGN: A prospective study. OBJECTIVE: To investigate the association between microstructural changes measured by diffusion tensor imaging (DTI) and clinical symptoms and their duration in patients with cervical spondylotic myelopathy (CSM) affected by single level. SUMMARY OF BACKGROUND DATA: No report was reported regarding the association between the microstructural changes and the symptoms and their duration at single-level spinal cord compression. METHODS: Twenty-nine consecutive patients with CSM and 29 normal subjects were enrolled in this study. DTI with tractography was performed on the cervical spinal cord. Clinical symptoms were evaluated using modified Japanese Orthopaedic Association (mJOA) scores for each patient, and the duration of clinical symptoms was noted based on the earliest instance of limb pain or numbness or weakness or bladder dysfunction. Mean fractional anisotropy (FA) and apparent diffusion coefficient (ADC) values were calculated from tractography images. RESULTS: The mean FA value of the cervical compressed spinal cord was significantly lower than the FA of the normal population (P < 0.001). The mean ADC value in the cervical compressed spinal cord was obviously higher than those of normal cervical spinal cord (P < 0.001). In the CSM patients, a significant positive association was observed between FA values and mJOA scores (P < 0.001). However, there were a notable negative association between mJOA scores and ADC values (P < 0.001), and between mJOA scores and symptom duration (P < 0.001). CONCLUSION: These results illustrate DTI can measure the micostructural changes of cervical spinal cord and DTI parameters are potential biomarkers for spinal cord dysfunction in patients with CSM. LEVEL OF EVIDENCE: 3.

The Role of Gene Expression Changes in ceRNA Network Underlying Ossification of Ligamentum Flavum Development.

We aimed at exploring the role of gene expression changes regulated by non-coding RNAs in ossification of ligamentum flavum (OLF). Three microarray datasets, including long non-coding RNA (lncRNA)/mRNA expression profile (GSE106253), circular RNA (circRNA) expression profile (GSE106255), and microRNA (miRNA) expression profile (GSE106256), were downloaded from the public Gene Expression Omnibus repository. The differentially expressed (DE) mRNAs, lncRNAs, miRNAs, and circRNAs in OLF tissues were analyzed, compared with normal tissues. Two competing endogenous RNA (ceRNA) networks with lncRNA-miRNA-mRNA and circRNA-miRNA-mRNA interactions were constructed, separately. Random walk with the restart model was applied to calculate the correlations of mRNAs with the published OLF-related genes. The top 50 mRNAs were subjected to function enrichment analysis and active small-molecule prediction. Total 2323 DE mRNAs, 1168 lncRNAs, 336 circRNAs, and 29 miRNAs were identified based on the microarray datasets. The LncRNA-related ceRNA network was constructed with 614 lncRNA-miRNA, 494 miRNA-mRNA, and 2099 lncRNA-mRNA interaction pairs; the circRNA-related ceRNA network was constructed with 153 circRNA-miRNA, 190 miRNA-mRNA, and 210 circRNA-mRNA interaction pairs. There were 17 OLF-related genes retrieved from previous literature, such as NPPS, COL6A1, and COL11A2, among which COL6A1 was the overlapped gene with mRNAs in the ceRNA network. Subsequently, top 50 mRNAs that closely correlated with COL6A1 in the ceRNA network were captured and these genes were closely related with the collagen catabolic process, regulation of cell growth, and neuronal action potential. DRD1 and COL6A1 were predicted to be the targets by small active molecule drugs. The collagen catabolic process may be implicated in OLF development. COL6A1 and DRD1 may be the candidate targets for OLF. However, further validations were needed.

High glucose mediates NLRP3 inflammasome activation via upregulation of ELF3 expression.

Microtubule affinity regulating kinase 4 (MARK4) plays a crucial role in the regulation of NOD-like receptor pyrin domain 3 (NLRP3) inflammasome activation, which leads to the generation of bioactive interleukin (IL)-1ß and IL-18. E74-like ETS transcription factor 3 (ELF3) participates in endothelial inflammatory processes. We hypothesized that ELF3 modulates MARK4 expression in vascular endothelial cells, thus contributing to high glucose-mediated NLRP3 inflammasome activation. Plasma IL-1ß, IL-18, NLRP3 inflammasome and MARK4 expression was increased in diabetic patients and rats. An in vitro study indicated that high glucose increased IL-1ß and IL-18 expression and activated the NLRP3 inflammasome via upregulation of MARK4 in human umbilical vein endothelial cells (HUVECs). Furthermore, high glucose increased ELF3 expression. ELF3 downregulation reversed the effects of high glucose treatment. Accordingly, the effects of ELF3 overexpression were similar to those of high glucose treatment and were counteracted by siMARK4. Furthermore, ELF3 was found to interact with SET8. High glucose inhibited SET8 expression and histone H4 lysine 20 methylation (H4K20me1), a downstream target of SET8. Overexpression of SET8 inhibited high glucose-induced MARK4 expression and NLRP3 inflammasome activation. The effects of shSET8 were similar to those of high glucose treatment and were counteracted by siMARK4. A mechanistic study found that ELF3 and H4K20me1 were enriched in the MARK4 promoter region. si-ELF3 attenuated MARK4 promoter activity and augmented the inhibitory effect of SET8 on MARK4 promoter activity. Furthermore, SET8 downregulation and ELF3 upregulation were confirmed in diabetic patients and rats. In conclusion, ELF3 interacted with SET8 to modulate MARK4 expression, which participated in hyperglycaemia-mediated endothelial NLRP3 inflammasome activation.

Electric Field-Induced Assembly and Alignment of Silver-Coated Cellulose for Polymer Composite Films with Enhanced Dielectric Permittivity and Anisotropic Light Transmission.

Multifarious wearable electronics with flexible touch screens have been invented for extensive outdoor activities. One challenge associated with these wearable electronics is the development of materials with both high dielectric permittivity and anisotropic light transmission, which is responsible for high touch sensitivity and screen peep-proof protection, respectively. Herein, we demonstrated a scalable approach for assembling and aligning anisotropic cellulose in a polymer matrix through the thickness direction via the assistance of an electric field to address this challenge. The alignment of silver-coated fibrillated celluloses in the polymer matrix not only significantly increases dielectric permittivity but also effectively enhances optical anisotropy. The impact of alignment degree and filler content on the dielectric and optical properties of polymer composite films has been systematically studied. The kinetics and aligning mechanisms of silver-coated fibrillated celluloses are revealed by in situ optical microscope images while an electric field is applied. We believe that this study provides a facile strategy to enhance both dielectric permittivity and optical anisotropy of polymer composite films by the alignment of embedding nanoparticles via an AC electric field, which is essential for future flexible electronics and display technology.

Correction: Homology analysis of malware based on ensemble learning and multifeatures.

[This corrects the article DOI: 10.1371/journal.pone.0211373.].

Multi-Source Deep Transfer Neural Network Algorithm.

Transfer learning can enhance classification performance of a target domain with insufficient training data by utilizing knowledge relating to the target domain from source domain. Nowadays, it is common to see two or more source domains available for knowledge transfer, which can improve performance of learning tasks in the target domain. However, the classification performance of the target domain decreases due to mismatching of probability distribution. Recent studies have shown that deep learning can build deep structures by extracting more effective features to resist the mismatching. In this paper, we propose a new multi-source deep transfer neural network algorithm, MultiDTNN, based on convolutional neural network and multi-source transfer learning. In MultiDTNN, joint probability distribution adaptation (JPDA) is used for reducing the mismatching between source and target domains to enhance features transferability of the source domain in deep neural networks. Then, the convolutional neural network is trained by utilizing the datasets of each source and target domain to obtain a set of classifiers. Finally, the designed selection strategy selects classifier with the smallest classification error on the target domain from the set to assemble the MultiDTNN framework. The effectiveness of the proposed MultiDTNN is verified by comparing it with other state-of-the-art deep transfer learning on three datasets.

Homology analysis of malware based on ensemble learning and multifeatures.

With the exponential increase in malware, homology analysis has become a hot research topic in the malware detection field. This paper proposes MHAS, a malware homology analysis system based on ensemble learning and multifeatures. MHAS generates grayscale images from malware binary files and then uses the opcode tool IDA Pro to extract opcode sequences and system call graphs. Thus, RGB images and M-images are generated on the image matrix. Then, MHAS uses convolutional neural networks (CNNs) as base learners to perform bagging ensemble learning to learn features from the grayscale images, RGB images and M-images. Next, MHAS integrates the nine base learners using voting, learning and selective ensemble (in that order) and maps the integration results to the result matrix. Finally, the result matrix is again integrated using the learning method to obtain the final malware classification result. To verify the accuracy of MHAS, we performed a malware family classification experiment, that included samples of 10 malware families. The results showed that MHAS can reach an accuracy rate of 99.17%, meaning that it can effectively analyze and identify malware families.