Search

COVID-19 has triggered 6.42 million death tolls, and more than 586 million confirmed positive cases until 10/Aug/2022. CT-based diagnosis methods need special expert knowledge, and the labeling procedure is tedious. We first propose a 12-layer CNN-based backbone network. Then, we utilize the Swish activation function to replace traditional ReLU. The multiple-way data augmentation is utilized to enhance the training set. Our model is named Swish-based CNN (SCNN). A web app is developed based on the proposed SCNN model. The SCNN model performs better than the ReLU-based backbone network and LReLU-based backbone network, indicating the effectiveness of the Swish function. The SCNN model achieves a sensitivity of 94.50 ± 1.06, a specificity of 95.25 ± 0.59, and an accuracy of 94.88 ± 0.65. It performs better than ten state-of-the-art COVID-19 diagnosis methods. Our SCNN model is promising in diagnosing COVID-19. The developed web app can help the users upload their own images and give the prediction results. [ABSTRACT FROM AUTHOR]

COVID-19 has caused over 6.35 million deaths and over 555 million confirmed cases till 11/July/2022. It has caused a serious impact on individual health, social and economic activities, and other aspects. Based on the gray-level co-occurrence matrix (GLCM), a four-direction varying-distance GLCM (FDVD-GLCM) is presented. Afterward, a five-property feature set (FPFS) extracts features from FDVD-GLCM. An extreme learning machine (ELM) is used as the classifier to recognize COVID-19. Our model is finally dubbed FECNet. A multiple-way data augmentation method is utilized to boost the training sets. Ten runs of tenfold cross-validation show that this FECNet model achieves a sensitivity of 92.23 ± 2.14, a specificity of 93.18 ± 0.87, a precision of 93.12 ± 0.83, and an accuracy of 92.70 ± 1.13 for the first dataset, and a sensitivity of 92.19 ± 1.89, a specificity of 92.88 ± 1.23, a precision of 92.83 ± 1.22, and an accuracy of 92.53 ± 1.37 for the second dataset. We develop a mobile app integrating the FECNet model, and this web app is run on a cloud computing-based client–server modeled construction. This proposed FECNet and the corresponding mobile app effectively recognize COVID-19, and its performance is better than five state-of-the-art COVID-19 recognition models. [ABSTRACT FROM AUTHOR]

The rampant spread of explicit content across social media can leave a damaging mark on our society. Hence, the need to be vigilant in detecting and curtailing sexually explicit content cannot be overstated. As such, it becomes paramount to discern and manage sexually explicit material to curb its dissemination and safeguard our digital communities from its harmful effects. In this article, we propose a unique technique entitled attention‐enabled pooling (ABP) embedded Swin transformer‐based YOLOv3 (ASYv3) for the detection of obscene areas present in the images with a bounding box around the offensive regions. ASYv3 employs a unique two‐step approach for enhanced performance in obscene detection. In the first step, a scalable and efficient Swin transformer block is integrated, utilizing self‐attention and model parallelism to train massive models effectively. In the second phase, the embedding layer of the Swin transformer is replaced with ABP, mitigating disruption of feature context. ABP allows for the projection of raw‐valued features into linear form with proper attention to feature context information at specified locations, resulting in optimized feature extraction. The proposed ABP embedded Swin transformer‐based YOLOv3 (ASYv3) was trained with annotated obscene images (AOI) dataset. The proposed ASYv3 model surpassed the state‐of‐the‐art methods by achieving 97% testing accuracy, 96.62% precision, 97.40% sensitivity, 3.48% FPR rate, 97.37% NPV values, and 95.59% mAP values, respectively. [ABSTRACT FROM AUTHOR]

Countless cybercrime instances have shown the need for detecting and blocking obscene material from social media sites. Deep learning methods (DLMs) outperformed in recognizing obscene content flooded on many online platforms. However, these contemporary DLMs primarily treat the recognition of obscene content as a simple task of binary classification, rather than focusing on the labelling of obscene areas. Hence, many of these methods could not pay attention to the fact that misclassification samples are so diverse. Therefore, this paper focuses on two aspects (i) developing a deep learning model that could classify and label the obscene portion, and (ii) generating a labelled obscene image dataset with a wide variety of obscene samples to minimize the risks of inaccurate recognition. We have proposed a method named S3Pooling based bottleneck attention module (BAM) embedded MobileNetV2‐YOLOv3 (SBMYv3) for automatic detection of obscene content using an attention mechanism and a suitable pooling strategy. The key contributions of our article are: (i) generation of a well‐labelled obscene image dataset with a variety of augmentation strategies using Pix‐2‐Pix GAN (ii) modifications to the backend architecture of YOLOv3 using MobileNetV2 and BAM to ensure focused and accurate feature extraction, and (iii) selection of an optimal pooling strategy, that is, S3Pooling strategy, while taking the design of the feature extractor into account. The proposed SBMYv3 model outperformed other state‐of‐the‐art models with 99.26% testing accuracy, 99.39% recall, 99.13% precision, and 99.13% IoU values respectively. [ABSTRACT FROM AUTHOR]

As a hearing disorder, sensorineural hearing loss (SNHL) can be effectively detected by magnetic resonance imaging (MRI). However, the manual detection of MRI scanning is subjective, time‐consuming, and unpredictable. An accurate and automatic computer‐aided diagnosis system is proposed for SNHL detection, providing reliable references for professionals. The system first employs a wavelet entropy layer to extract features of MRI images. Then, a neural network layer is proposed as the classifier consisting of a feedforward neural network (FNN) and an adaptive‐probability PSO (APPSO) algorithm. The authors prove the rotation‐variant property of the basic particle swarm optimization (PSO) by the algebraic property of matrix transformation. The property is unsuitable for optimising parameters of neural networks. Thus, in APPSO, the authors integrate the new update rules based on all‐dimensional variation and adaptive‐probability mechanism into the basic PSO, which can improve its searching ability without losing population diversity. The authors compare APPSO‐NN with FNN trained by five popular evolutionary algorithms. The simulation results show that APPSO performs best in training FNN. The method also compares with six state‐of‐the‐art methods. The simulation results show that the best performance in sensitivity and overall accuracy of hearing loss classification, which proves that the method is effective and promising for SNHL detection. [ABSTRACT FROM AUTHOR]

Ophiocordyceps sinensis (syn. Cordyceps sinensis) is a valuable medicinal fungus in traditional Chinese medicine, and one or more polysaccharides are the key constituents with important medical effects. Glycogen as a functional polysaccharide is widely identified in eukaryotes including fungi. However, there is no definitive report of glycogen presence in O. sinensis. In this study, we carefully fractionated polysaccharides from cultivated caterpillar fungus O. sinensis , which were then characterized via methods for glycogen analysis. According to the results, 1.03 ± 0.43 % of polysaccharides were quantified via amyloglucosidase digestion in the whole cultivated caterpillar fungus, which had a typical spherical shape under transmission electron microscope with an average peak radius of 37.63 ± 0.57 nm via size exclusion chromatography and an average chain length of 12.47 ± 0.94 degree of polymerization via fluorophore-assisted capillary electrophoresis. Taken together, this study confirmed that the polysaccharides extracted form O. sinensis were mostly glycogen. • Cultivated caterpillar fungus O. sinensis stores 1.03 ± 0.43 % of glycogen. • Spherical glycogen particles were identified with peak radius at 37.63 ± 0.57 nm. • Average chain length of O. sinensis glycogen was 12.47 ± 0.94 DP. [ABSTRACT FROM AUTHOR]

Aim: Fruit category classification is important in factory packing and transportation, price prediction, dietary intake, and so forth. Methods: This study proposed a novel artificial intelligence system to classify fruit categories. First, 2D fractional Fourier entropy with rotation angle vector grid was used to extract features from fruit images. Afterwards, a five‐layer stacked sparse autoencoder was used as the classifier. Results: Ten runs on the test set showed our method achieved a micro‐averaged F1 score of 95.08% for an 18‐category fruit dataset. Conclusion: Our method gives better micro‐averaged F1 score than 10 state‐of‐the‐art approaches. [ABSTRACT FROM AUTHOR]

Within the framework of nonrelativistic QCD (NRQCD) factorization, we calculate the next-to-leading-order (NLO) perturbative corrections to the radiative decay Υ → ηc(χcJ) + γ. Both the helicity amplitudes and the helicity decay widths are obtained. It is the first computation for the processes involving both bottomonium and charmonium at two-loop accuracy. By employing the Cheng-Wu theorem, we are able to convert most of complex-valued master integrals (MIs) into real-valued MIs, which makes the numerical integration much efficient. Our results indicate the O α s corrections are moderate for ηc and χc2 production, and are quite marginal for χc0 and χc1 production. It is impressive to note the NLO corrections considerably reduce the renormalization scale dependence in both the decay widths and the branching fractions for χcJ, and slightly improve that for ηc. With the NRQCD matrix elements evaluated via the Buchmüller-Tye potential model, we find the decay width for ηc production is one-order-of-magnitude larger than χcJ production, which may provide a good opportunity to search for Υ → ηc + γ in experiment. In addition, the decay width for χc1 production is several times larger than those for χc0,2. Finally, we find the NLO NRQCD prediction for the branching fraction of Υ → χc1 + γ is only half of the lower bound of the experimental data measured recently by Belle. Moreover, there exists serious contradiction between theory and experiment for Υ → ηc + γ. The discrepancies between theory and experiment deserve further research efforts. [ABSTRACT FROM AUTHOR]

• Convolutional block attention module is included in the proposed model. • The proposed multiple-input end-to-end model can handle CCT and CXR images simultaneously. • Multiple-way data augmentation is employed to overcome overfitting problem. • The proposed model gives more accurate performances than individual modality. • The proposed model offers better performances than state-of-the-art approaches. COVID-19 has caused 3.34m deaths till 13/May/2021. It is now still causing confirmed cases and ongoing deaths every day. This study investigated whether fusing chest CT with chest X-ray can help improve the AI's diagnosis performance. Data harmonization is employed to make a homogeneous dataset. We create an end-to-end multiple-input deep convolutional attention network (MIDCAN) by using the convolutional block attention module (CBAM). One input of our model receives 3D chest CT image, and other input receives 2D X-ray image. Besides, multiple-way data augmentation is used to generate fake data on training set. Grad-CAM is used to give explainable heatmap. The proposed MIDCAN achieves a sensitivity of 98.10±1.88%, a specificity of 97.95±2.26%, and an accuracy of 98.02±1.35%. Our MIDCAN method provides better results than 8 state-of-the-art approaches. We demonstrate the using multiple modalities can achieve better results than individual modality. Also, we demonstrate that CBAM can help improve the diagnosis performance. [ABSTRACT FROM AUTHOR]

• This work proposes a Modified VGG19 deep-learning architecture to diagnose the Chest X-Rays. • This work computes the essential handcrafted features from the Chest X-Rays. • This work introduces an Ensemble of Features Scheme (EFS) by integrating the deep-features and the handcrafted features. • Serial fusion and PCA based selection is implemented in EFS to choose primary feature set. • Experimental results demonstrate superior performance of Modified VGG19 in comparison with AlexNet, VGG16, VGG19 and ResNet50. Pneumonia is one of the major illnesses in children and aged humans due to the Infection in the lungs. Early analysis of pneumonia is necessary to prepare for a possible treatment procedure to regulate and cure the disease. This research aspires to develop a Deep-Learning System (DLS) to diagnose the lung abnormality using chest X-ray (radiograph) images. The proposed work is implemented using; (i) Conventional chest radiographs and (ii) Chest radiograph treated with a threshold filter. The initial experimental evaluation is carried out using the traditional DLS, such as AlexNet, VGG16, VGG19 and ResNet50 with a SoftMax classifier. The results confirmed that, VGG19 provides better classification accuracy (86.97%) compared to other methods. Later, a customized VGG19 network is proposed using the Ensemble Feature Scheme (EFS), which combines the handcrafted features attained with CWT, DWT and GLCM with the Deep-Features (DF) achieved using Transfer-Learning (TL) practice. The performance of customized VGG19 is tested using different classifiers, such as SVM-linear, SVM-RBF, KNN classifier, Random-Forest (RF) and Decision-Tree (DT). The result confirms that VGG19 with RF classifier offers better accuracy (95.70%). When the similar experiment is repeated using threshold filter treated chest radiographs, the VGG19 with RF classifier offered superior classification accuracy (97.94%). This result confirms that, proposed DLS will work well on the benchmark images and in the future, it can be considered to diagnose clinical grade chest radiographs. To create your abstract, type over the instructions in the template box below. Fonts or abstract dimensions should not be changed or altered. Image, graphical abstract [ABSTRACT FROM AUTHOR]

Till August 17, 2020, COVID-19 has caused 21.59 million confirmed cases in more than 227 countries and territories, and 26 naval ships. Chest CT is an effective way to detect COVID-19. This study proposed a novel deep learning model that can diagnose COVID-19 on chest CT more accurately and swiftly. Based on traditional deep convolutional neural network (DCNN) model, we proposed three improvements: (i) We introduced stochastic pooling to replace average pooling and max pooling; (ii) We combined conv layer with batch normalization layer and obtained the conv block (CB); (iii) We combined dropout layer with fully connected layer and obtained the fully connected block (FCB). Our algorithm achieved a sensitivity of 93.28% ± 1.50%, a specificity of 94.00% ± 1.56%, and an accuracy of 93.64% ± 1.42%, in identifying COVID-19 from normal subjects. We proved using stochastic pooling yields better performance than average pooling and max pooling. We compared different structure configurations and proved our 3CB + 2FCB yields the best performance. The proposed model is effective in detecting COVID-19 based on chest CT images. [ABSTRACT FROM AUTHOR]

Aim: Workplace social capital refers to relationship networks formed by individuals in an organization through long‐term mutually beneficial interactions and cooperation with members. These relationship networks can create value and resources for organizations and individuals. This current study aimed to explore the potential impact of workplace social capital on the association between perceived stress and professional identity in clinical nurses during the COVID‐19 outbreak. Methods: In this cross‐sectional study, 308 Chinese clinical nurses filled out the Chinese Workplace Social Capital Scale, the Chinese Perceived Stress Scale, and the Chinese Nurse's Professional Identity Scale. Descriptive analysis, independent samples t test, analysis of variance, Pearson correlation analyses, and bootstrap method were performed to analyze the data. Results: Perceived stress was negatively correlated with professional identity (r = −0.455, p <.001). Workplace social capital was not found to moderate the relationship between perceived stress and professional identity (95% CI −0.03 to− 0.06, p =.47 >.05). Instead, it mediated that relationship (95% CI −0.61 to −0.19, p <.05), and its mediating effect was −0.37. Conclusions: In the early stages of the COVID‐19 outbreak, workplace social capital among the investigated clinical nurses failed to buffer the negative impact of perceived stress on professional identity, but it did play a part in mediating perceived stress and professional identity. A healthy workplace should be provided to clinical nurses to improve their professional identity, while lowering perceived stress. [ABSTRACT FROM AUTHOR]

• We analysed over 450 references from all well-famed databases. • We provided a comprehensive survey on multimodal data fusion in neuroimaging. • This review encompassed current challenges & applications, strengths &limitations. • Fundamental fusion rules, and fusion quality assessment methods were reviewed. • Atlas-based fusion segmentation, quantification, & applications were reviewed. Multimodal fusion in neuroimaging combines data from multiple imaging modalities to overcome the fundamental limitations of individual modalities. Neuroimaging fusion can achieve higher temporal and spatial resolution, enhance contrast, correct imaging distortions, and bridge physiological and cognitive information. In this study, we analyzed over 450 references from PubMed, Google Scholar, IEEE, ScienceDirect, Web of Science, and various sources published from 1978 to 2020. We provide a review that encompasses (1) an overview of current challenges in multimodal fusion (2) the current medical applications of fusion for specific neurological diseases, (3) strengths and limitations of available imaging modalities, (4) fundamental fusion rules, (5) fusion quality assessment methods, and (6) the applications of fusion for atlas-based segmentation and quantification. Overall, multimodal fusion shows significant benefits in clinical diagnosis and neuroscience research. Widespread education and further research amongst engineers, researchers and clinicians will benefit the field of multimodal neuroimaging. [ABSTRACT FROM AUTHOR]

Sitodiplosis mosellana, a periodic but devastating wheat pest, relies on wheat spike volatiles as a cue in selecting hosts for oviposition. Insect odorant‐binding proteins (OBPs) are thought to play essential roles in filtering, binding and transporting hydrophobic odorant molecules to specific receptors. To date, the molecular mechanisms underlying S. mosellana olfaction are poorly understood. Here, three S. mosellana antenna‐specific OBP genes, SmosOBP11, 16 and 21, were cloned and bacterially expressed. Binding properties of the recombinant proteins to 28 volatiles emitted from wheat spikes were investigated using fluorescence competitive binding assays. Sequence analysis suggested that these SmosOBPs belong to the Classic OBP subfamily. Ligand‐binding analysis showed that all three SmosOBPs preferentially bound alcohol, ester and ketone compounds, and SmosOBP11 and 16 also selectively bound terpenoid compounds. In particular, the three SmosOBPs had high binding affinities (Ki < 20 μmol/L) to 3‐hexanol and cis‐3‐hexenylacetate that elicited strong electroantennogram (EAG) response from female antennae. In addition, SmosOBP11 displayed significantly higher binding (Ki < 8 μmol/L) than SmosOBP16 and 21 to 1‐octen‐3‐ol, D‐panthenol, α‐pinene and heptyl acetate which elicited significant EAG response, suggesting that SmosOBP11 plays a major role in recognition and transportation of these volatiles. These findings have provided important insight into the molecular mechanism by which S. mosellana specifically recognizes plant volatiles for host selection, and have facilitated identification of effective volatile attractants that are potentially useful for pest monitoring and trapping. [ABSTRACT FROM AUTHOR]

AIM: Unilateral sensorineural hearing loss is a brain disease, which causes slight morphology changes within brain structure. Traditional manual method may ignore this change. METHOD: In this work, we developed a novel method, based on the double-density dual-tree complex (DDDTCWT), and radial basis function kernel principal component analysis (RKPCA) and multinomial logistic regression (MLR) for the magnetic resonance imaging scanning. We first used DDDTCWT to extract features. Afterwards, we used RKPCA to reduce feature dimensionalities. Finally, MLR was employed to be the classifier. RESULT: The 10 times of 10-fold stratified cross validation showed our method achieved an overall accuracy of 96.44 ± 0.88%. The sensitivities of detecting left-sided sensorineural hearing loss, right-sided sensorineural hearing loss, and healthy controls were 96.67 ± 2.72%, 96.67 ± 3.51%, and 96.00 ± 4.10%, respectively. CONCLUSION: Our method performed better than both raw and improved AlexNet, and eight state-of-the-art methods via a stringent statistical 10 × 10-fold stratified cross validation. The MLR gives better classification performance than decision tree, support vector machine, and back-propagation neural network. [ABSTRACT FROM AUTHOR]

Objectives: To compare a multi-feature-based radiomic biomarker with volumetric analysis in discriminating lung adenocarcinomas with different disease-specific survival on computed tomography (CT) scans.Methods: This retrospective study obtained institutional review board approval and was Health Insurance Portability and Accountability Act (HIPAA) compliant. Pathologically confirmed lung adenocarcinoma (n = 431) manifested as subsolid nodules on CT were identified. Volume and percentage solid volume were measured by using a computer-assisted segmentation method. Radiomic features quantifying intensity, texture and wavelet were extracted from the segmented volume of interest (VOI). Twenty best features were chosen by using the Relief method and subsequently fed to a support vector machine (SVM) for discriminating adenocarcinoma in situ (AIS)/minimally invasive adenocarcinoma (MIA) from invasive adenocarcinoma (IAC). Performance of the radiomic signatures was compared with volumetric analysis via receiver-operating curve (ROC) analysis and logistic regression analysis.Results: The accuracy of proposed radiomic signatures for predicting AIS/MIA from IAC achieved 80.5% with ROC analysis (Az value, 0.829; sensitivity, 72.1%; specificity, 80.9%), which showed significantly higher accuracy than volumetric analysis (69.5%, P = 0.049). Regression analysis showed that radiomic signatures had superior prognostic performance to volumetric analysis, with AIC values of 81.2% versus 70.8%, respectively.Conclusions: The radiomic tumour-phenotypes biomarker exhibited better diagnostic accuracy than traditional volumetric analysis in discriminating lung adenocarcinoma with different disease-specific survival.Key Points: • Radiomic biomarker on CT was designed to identify phenotypes of lung adenocarcinoma • Built up radiomic signature for lung adenocarcinoma manifested as subsolid nodules • Retrospective study showed radiomic signature had greater diagnostic accuracy than volumetric analysis • Radiomics help to evaluate intratumour heterogeneity within lung adenocarcinoma • Medical decision can be given with more confidence. [ABSTRACT FROM AUTHOR]

Purpose: To investigate whether triphasic CT with a simplified Patlak plot can be used in clinical practice for the estimate of split kidney glomerular filtration rate (SKGFR). Materials and methods: The animal experiment included 15 rabbits that underwent 40 dynamic contrast-enhanced CT scans of the kidneys with 1.5 s time interval. Patlak-derived SKGFR was obtained using standard forty-point, two-point (unenhanced phase, arterial phase t and portovenous phase t ), and a modified two-point (MTP) (unenhanced, t , t , and a virtual t [ t = ( t + t )/2]) image data, respectively. The MTP-Patlak plot approach was then validated in 13 patients who underwent a triphasic renal contrast-enhanced CT examination. SKGFR measured by 99Tc-DTPA clearance was as a standard reference. Results: MTP-Patlak significantly reduced input function errors than two-point Patlak (21.1 ± 16.2 % vs 30.8 ± 15.2 %, p < 0.01) and showed good concordance with standard Patlak for measurement of SKGFR in animal experiment (1.20 ± 0.38 mL/g/min vs 1.51 ± 0.43 mL/g/min; linear correlation coefficient r = 0.87, p < 0.001). Human study showed that mean SKGFR was 45.7 mL/min (range, 26.5-86.2 mL/min) obtained from 99Tc-DTPA, and 38.2 mL/min (range, 18.6-79.3 mL/min) obtained from triphasic CT using MTP-Patlak plot. Linear correlation between the two methods was r = 0.75 ( p < 0.01). The mean difference between SKGFRs as determined with the two methods was 7.4 ± 9.0 mL/min. Conclusion: The MTP-Patlak approach, featured with simplicity, is feasible in a clinically indicated CT examination for the evaluation of split renal function. [ABSTRACT FROM AUTHOR]

The early diagnosis of skin cancer through clinical methods reduces the human mortality rate. The manual screening of dermoscopic images is not an efficient procedure; therefore, researchers working in the domain of computer vision employed several algorithms to classify the skin lesion. The existing computerized methods have a few drawbacks, such as low accuracy and high computational time. Therefore, in this work, we proposed a novel deep learning and Entropy‐Normal Distribution Optimization Algorithm with extreme learning machine (NDOEM)‐based architecture for multiclass skin lesion classification. The proposed architecture consists of five fundamental steps. In the first step, two contrast enhancement techniques including hybridization of mathematical formulation and convolutional neural network are implemented prior to data augmentation. In the second step, two pre‐trained deep learning models, EfficientNetB0 and DarkNet19, are fine‐tuned and retrained through the transfer learning. In the third step, features are extracted from the fine‐tuned models and later the most discriminant features are selected based on novel Entropy‐NDOELM algorithm. The selected features are finally fused using a parallel correlation technique in the fourth step to generate the result feature vectors. Finally, the resultant features are again down‐sampled using the proposed algorithm and the resultant features are passed to the extreme learning machine (ELM) for the final classification. The simulations are conducted on three publicly available datasets as HAM10000, ISIC2018, and ISIC2019 to achieving an accuracy of 95.7%, 96.3%, and 94.8% respectively. [ABSTRACT FROM AUTHOR]

ABSTRACT To classify brain images into pathological or healthy is a key pre-clinical state for patients. Manual classification is tiresome, expensive, time-consuming, and irreproducible. In this study, we aimed to present an automatic computer-aided system for brain-image classification. We used 90 T2-weighted images obtained by magnetic resonance images. First, we used weighted-type fractional Fourier transform (WFRFT) to extract spectrums from each magnetic resonance image. Second, we used principal component analysis (PCA) to reduce spectrum features to only 26. Third, those reduced spectral features of different samples were combined and were fed into support vector machine (SVM) and its two variants: generalized eigenvalue proximal SVM and twin SVM. A 5 × 5-fold cross-validation results showed that this proposed 'WFRFT + PCA + generalized eigenvalue proximal SVM' yielded sensitivity of 99.53%, specificity of 92.00%, precision of 99.53%, and accuracy of 99.11%, which are comparable with the proposed 'WFRFT + PCA + twin SVM' and better than the proposed 'WFRFT + PCA + SVM.' Besides, all three proposed methods were superior to eight state-of-the-art algorithms. Thus, WFRFT is effective, and the proposed methods can be used in practical. © 2015 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 25, 317-327, 2015 [ABSTRACT FROM AUTHOR]

• Specify the significance of multimodality data fusion in neuroimaging. • Summarize the challenges of multimodality data fusion in neuroimaging. • Provide recent research advances in multimodality data fusion in neuroimaging. [ABSTRACT FROM AUTHOR]

Osteoporosis (OP) is an osteometabolic disorder characterized by a lesser bone mineral density (BMD) and the disruption of bone tissue micro—architecture, resulting in a greater bone fragility and higher likelihoods of fractures. OP emerges once the bone mass declines faster than the body's capacity to replenish it, leading to a significant reduction in the strength of bone. OP impacts each and every bone throughout the body and provides no clinical signs until a fracture happens. Aging leads to the reduction in BMD, and the rate of percentage of fractures rises over time, usually causing mortality and morbidity. Numerous BMD evaluation methods are available, and they are used in a variety of settings by considering the location of the fracture. Dual-energy x-ray absorptiometry (DEXA/DXA) is recognized as the gold standard for predicting the fracture, since it is the most advanced, technologically affirmed, and also has excellent performance. According to the findings, most researchers do not endeavour the identification and the segmentation of low bone masses from DEXA images. Medical image segmentation supports in analyzing and visualizing the bone's low bone mass. The envisaged hybrid approach, that integrates GLCM for feature extraction and AlexNet for a low bone mass variation classification, provides segmented images that assist in categorizing bone health as normal, osteopenia, or osteoporosis. The developed algorithm's performance metrics, including Dice Co-efficient, Sensitivity, and Specificity, were 92.35, 90.26, and 92.42%, respectively. The Orthopedicians ascertained the efficacy of the outcomes rendered by the proposed algorithm. [ABSTRACT FROM AUTHOR]

Dynamic contrast enhanced (DCE) MR renography has been identified as an interesting tool to determine single-kidney GFR. However, a fundamental issue for the applicability of MR-based estimate of single-kidney GFR is selecting a balance between spatial and temporal resolution of DCE-MRI data. The purpose is to assess the feasibility of GFR estimate from high-resolution (HR) dynamic contrast-enhanced (DCE) MRI in a routine clinical modal. Standard MR renography (2.4 s/phase, total 4 min; 4-ml Gd) and five-phase, HR-based imaging protocol (0, 30, 70, 120, and 240 s; 0.05 mmol/kg Gd) were prospectively performed in twelve volunteers who were scheduled for routine renal MRI. Data were plotted with Patlak, two-compartment modified Tofts model (2CTM), and two-compartment filtration model (2CFM) for GFR estimate. During all the measurements, only the signal intensities in the aorta and whole kidney parenchyma were considered. Standard 2CFM and 2CTM produced lower residuals over the fitted interval than HR-based measures ( p < 0.05); and HR-bases 2CFM and 2CTM did not reflect significant correlation to standard values. Standard Patlak plots with 0–240 s data points produced significantly lower GFR and higher residuals than that plots with 0–120 s data points ( p < 0.05). HR-based Patlak plots with 0, 30, 70, and 120 s data points significantly correlated with reference values (Pearson ρ = 0.97, p < 0.01), and produced a 33.2% underestimation of reference value, which was better than that plots with 0, 30, 70, 120, and 240 s data points ( ρ = 0.92, p < 0.01; 58.6% underestimation of reference value). It concludes that it is feasible to estimate GFR with HR-based DCE-MRI and appreciate kinetic model. Patlak plots from 0, 30, 70, and 120 s data points is better than plots from 0, 30, 70, 120, and 240 s data points. [ABSTRACT FROM AUTHOR]

Purpose To determine the relationship between myocardial structural and microvascular abnormality in hypertrophic cardiomyopathy (HCM) by multi-parametric cardiac MRI. Materials and methods Twenty-four HCM and eighteen controls were retrospectively included. Left ventricle mass (LVM), LV end-systolic and end-diastolic volume (LVESV, LVEDV), LV ejection fraction (LVEF), and 16-segment wall thickness at ES and ED (SESWT, SEDWT) were assessed with a 2D cine-MRI. Myocardial perfusion (reflected by K trans ), interstitial volume ( V e ) and mean transmit time (MTT) were evaluated with a model-dependent dynamic contrast-enhanced MRI. Myocardial fibrosis was assessed with late gadolinium enhancement (LGE) imaging. Results K trans was significantly decreased in LGE-present (0.74 ± 0.15 mL/g/min) against LGE-absent (0.55 ± 0.14 mL/g/min, p = 0.030) and normal group (0.81 ± 0.32 mL/g/min, p < 0.001), but was unchanged in LGE-absent against normal group ( p > 0.05). V e and MTT were significantly increased in LGE-present ( V e : 26.7 ± 15.7%; MTT: 28.6 ± 21.3 s) against LGE-absent (37.6 ± 18.3%; 49.8 ± 30.5 s) and normal group (19.7 ± 6.9%; 15.1 ± 3.9 s; all p < 0.001), and were significantly increased in LGE-absent against normal group ( p < 0.001). LGE significantly correlated to K trans , V e , MTT, and SESWT ( ρ = 0.232, −0.247, −0.443, and −0.207, respectively). K trans negatively correlated to SEDWT and SESWT ( ρ = −0.224 and −0.231). V e and MTT positively correlated to SEDWT ( V e : ρ = 0.223; MTT: ρ = 0.239) and SESWT ( V e : ρ = 0.248; MTT: ρ = 0.254). Conclusions Consistent relationship was determined between myocardial structural abnormality and microvascular dysfunction in HCM. [ABSTRACT FROM AUTHOR]

Objective: To evaluate histogram analysis of intravoxel incoherent motion (IVIM) for discriminating the Gleason grade of prostate cancer (PCa).Methods: A total of 48 patients pathologically confirmed as having clinically significant PCa (size > 0.5 cm) underwent preoperative DW-MRI (b of 0-900 s/mm(2)). Data was post-processed by monoexponential and IVIM model for quantitation of apparent diffusion coefficients (ADCs), perfusion fraction f, diffusivity D and pseudo-diffusivity D*. Histogram analysis was performed by outlining entire-tumour regions of interest (ROIs) from histological-radiological correlation. The ability of imaging indices to differentiate low-grade (LG, Gleason score (GS) ≤6) from intermediate/high-grade (HG, GS > 6) PCa was analysed by ROC regression.Results: Eleven patients had LG tumours (18 foci) and 37 patients had HG tumours (42 foci) on pathology examination. HG tumours had significantly lower ADCs and D in terms of mean, median, 10th and 75th percentiles, combined with higher histogram kurtosis and skewness for ADCs, D and f, than LG PCa (p < 0.05). Histogram D showed relatively higher correlations (ñ = 0.641-0.668 vs. ADCs: 0.544-0.574) with ordinal GS of PCa; and its mean, median and 10th percentile performed better than ADCs did in distinguishing LG from HG PCa.Conclusion: It is feasible to stratify the pathological grade of PCa by IVIM with histogram metrics. D performed better in distinguishing LG from HG tumour than conventional ADCs.Key Points: • GS had relatively higher correlation with tumour D than ADCs. • Difference of histogram D among two-grade tumours was statistically significant. • D yielded better individual features in demonstrating tumour grade than ADC. • D* and f failed to determine tumour grade of PCa. [ABSTRACT FROM AUTHOR]

Domain adaptation is crucial for transferring the knowledge from the source labeled CT dataset to the target unlabeled MR dataset in abdominal multi-organ segmentation. Meanwhile, it is highly desirable to avoid the high annotation cost related to the target dataset and protect the source dataset privacy. Therefore, we propose an effective source-free unsupervised domain adaptation method for cross-modality abdominal multi-organ segmentation without source dataset access. The proposed framework comprises two stages. In the first stage, the feature map statistics-guided model adaptation combined with entropy minimization is developed to help the top segmentation network reliably segment the target images. The pseudo-labels output from the top segmentation network are used to guide the style compensation network to generate source-like images. The pseudo-labels output from the middle segmentation network is used to supervise the learning progress of the desired model (bottom segmentation network). In the second stage, the circular learning and pixel-adaptive mask refinement are used to further improve the desired model performance. With this approach, we achieved satisfactory abdominal multi-organ segmentation performance, outperforming the existing state-of-the-art domain adaptation methods. The proposed approach can be easily extended to situations in which target annotation data exist. With only one labeled MR volume, the performance can be leveled with that of supervised learning. Furthermore, the proposed approach is proven to be effective for source-free unsupervised domain adaptation in reverse direction. [ABSTRACT FROM AUTHOR]

Reversible data hiding (RDH) is a widely studied research domain due to the extensive applications in medical image transmission, military image transmission, and cloud computing. In this paper, we propose a novel RDH scheme in encrypted images which is having a high embedding rate without affecting the bit error rate. To perform data hiding, all the non-overlapping pixel sub-arrays from the encrypted image will be considered in a predefined order, and one-bit from the secret message can be hidden in each of the pixel sub-arrays through a controlled pixel mapping technique. The pixels in a selected sub-array P will be split into two groups: odd pixels and even pixels according to the parity of the index of the pixels. All the even pixels will be replaced by a new pixel value based on a predefined mapping function to hide a secret bit 1. To embed bit-value 0, no need to do any modifications in the pixel sub-array P. A new correlation measure between adjacent pixels is introduced in this manuscript to extract the secret message bit and to restore the original image. All the simulations of the proposed scheme are done on the standard images from USC-SIPI image dataset. The results show that the proposed scheme achieved a very good embedding rate without affecting the bit error rate or image recoverability. • The proposed scheme uses an adaptive pixel mapping function on selected pixels. • The proposed framework can use any secure image encryption scheme. • The proposed reversible data hiding is independent of image encryption. • We observed a better embedding rate without compromising the bit error rate. • The proposed data hiding process is not affecting the image encryption efficiency. [ABSTRACT FROM AUTHOR]

This special issue begins with article focusing on evolutionary computation algorithm for health image processing. In the next article, the deep convolutional neural networks and evolutionary algorithms are converged to improve the e ciency and accuracy of image forgery detection. The next article presents multi-sensor data fusion algorithm to perform e cient data analysis of the healthcare applications. [Extracted from the article]

We report on the fabrication of a subnanometer-thick epitaxial PbS film on polydiacetylene (PDA) substrate. The film is formed by cation-absorption in PbCl2solution followed by sulfidation in mixed H2S/N2gas, exhibiting a flat and smooth surface morphology. The strong adsorbate−surface interaction between the lead hydrolysis complexes and the carboxyl headgroups on the surface of PDA accounts for such a growth mode. We further demonstrate that such an ultrathin film can be used as a buffer layer for both homo- and heteroepitaxial growth of sulfide films. The X-ray and electron diffraction results reveal that the (001) planes of PbS are parallel to the surface of PDA and the (110) planes of PbS are parallel to the (110) planes of PDA. [ABSTRACT FROM AUTHOR]

Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis. Our study investigates the recognition of secondary pulmonary (SPTB). A novel F3 model is proposed. The first F means using a four-direction varying-distance gray-level co-occurrence matrix (FDVDGLCM) to analyze the chest CT images; the second F means a five-property feature set (FPFS) from the FDVDGLCM results; the third F means fuzzy support vector machine (FSVM). Besides, a slight adaption of multiple-way data augmentation is used to boost the training set. The 10 runs of 10-fold cross-validation demonstrate that this F3 model achieves a sensitivity of 93.68% ± 1.75%, a specificity of 94.17% ± 1.68%, a precision of 94.17% ± 1.55%, an accuracy of 93.92% ± 1.05%, an F1 score of 93.91% ± 1.07%, an MCC of 87.88% ± 2.09%, and an FMI of 93.92% ± 1.06%. The AUC is 0.9624. The FSVM can give better performance than ordinary SVM. The proposed F3 model is superior to six state-of-the-art SPTB recognition models. [ABSTRACT FROM AUTHOR]

Fingerspelling is a method of spelling words via hand movements. This study aims to propose a novel fingerspelling recognition system. We use 1320 fingerspelling images in our dataset. Our method is based on the convolutional neural network (CNN) model. We propose a 12-layer CNN as the backbone. Particularly, stochastic pooling (SP) is used to help solve the problems caused by max pooling or average pooling. In addition, an improved 20-way data augmentation method is proposed to circumvent overfitting. Our method is dubbed CNNSP. The results show that our CNNSP method achieved a micro-averaged F1 (MAF) score of 90.04 ± 0.82%. In contrast, the MAFs of l2-pooling, average pooling, and max pooling are 86.21 ± 1.12%, 87.54 ± 1.39%, and 89.07 ± 0.78%, respectively. Our CNNSP attains better results than eight state-of-the-art fingerspelling recognition methods. Besides, the SP is better than l2-pooling, average pooling, and max pooling. [ABSTRACT FROM AUTHOR]

A Correction to this paper has been published: 10.1007/s13369-021-05881-4 [ABSTRACT FROM AUTHOR]

Purpose: The purpose of our study is to investigate image quality, efficiency, and diagnostic performance of a deep learning-accelerated single-shot breath-hold (DLSB) against BLADE for T2-weighted MR imaging (T2WI) for gastric cancer (GC). Methods: 112 patients with GCs undergoing gastric MRI were prospectively enrolled between Aug 2022 and Dec 2022. Axial DLSB-T2WI and BLADE-T2WI of stomach were scanned with same spatial resolution. Three radiologists independently evaluated the image qualities using a 5-scale Likert scales (IQS) in terms of lesion delineation, gastric wall boundary conspicuity, and overall image quality. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated in measurable lesions. T staging was conducted based on the results of both sequences for GC patients with gastrectomy. Pairwise comparisons between DLSB-T2WI and BLADE-T2WI were performed using the Wilcoxon signed-rank test, paired t-test, and chi-squared test. Kendall's W, Fleiss' Kappa, and intraclass correlation coefficient values were used to determine inter-reader reliability. Results: Against BLADE, DLSB reduced total acquisition time of T2WI from 495 min (mean 4:42 per patient) to 33.6 min (18 s per patient), with better overall image quality that produced 9.43-fold, 8.00-fold, and 18.31-fold IQS upgrading against BALDE, respectively, in three readers. In 69 measurable lesions, DLSB-T2WI had higher mean SNR and higher CNR than BLADE-T2WI. Among 71 patients with gastrectomy, DLSB-T2WI resulted in comparable accuracy to BLADE-T2WI in staging GCs (P > 0.05). Conclusions: DLSB-T2WI demonstrated shorter acquisition time, better image quality, and comparable staging accuracy, which could be an alternative to BLADE-T2WI for gastric cancer imaging. [ABSTRACT FROM AUTHOR]

• Modified Mask RCNN based skin lesion segmentation is performed. • 24-Layered CNN architecture is implemented. • Fully Connected layer is utilized for features mapping • Softmax classifier is employed for lesion type recognition A novel deep learning framework is proposed for lesion segmentation and classification. The proposed technique incorporates two primary phases. For lesion segmentation, Mask recurrent convolutional neural network (MASK R-CNN) based architecture is implemented. In this architecture, Resnet50 along with feature pyramid network (FPN) is utilized as a backbone. Later, fully connected layer-based features are mapped for the final mask generation. In the classification phase, 24-layered convolutional neural network architecture is designed, which performs activation based on the visualization of higher features. Finally, best CNN features are provided to softmax classifiers for final classification. Three datasets (i.e. PH2, ISBI2016, and ISIC2017) are utilized for the validation of the segmentation process, whilst HAM10000 dataset is utilized for the classification. From the results, it is concluded that the proposed method outperforms several existing techniques, based on the selected set of parameters including sensitivity (85.57%), precision (87.01%), F1- Score (86.28%), and accuracy (86.5%). [ABSTRACT FROM AUTHOR]

Due to the proliferation of biomedical imaging modalities, such as Photoacoustic Tomography, Computed Tomography (CT), Optical Microscopy and Tomography, etc., massive amounts of data are generated on a daily basis. While massive biomedical data sets yield more information about pathologies, they also present new challenges of how to fully explore the data. Data fusion methods are a step forward towards a better understanding of data by bringing multiple data observations together to increase the consistency of the information. However, data generation is merely the first step, and there are many other factors involved in the fusion process like noise, missing data, data scarcity, and high dimensionality. In this paper, an overview of the advances in data preprocessing in biomedical data fusion is provided, along with insights stemming from new developments in the field. [ABSTRACT FROM AUTHOR]

Forecasting the future movements of stock market indexes by utilizing historical transaction data is a prominent concern within the realm of finance. The application of graph convolutional networks to incorporate the interrelationships among various indices' patterns is a highly advanced subject within this field. Addressing the inconsistency between historical and future dynamic graphs in current graph convolution-based index prediction, we propose a method called G-Conv that constructs a graph structure based on constituent stocks of the indices for index trend prediction. This approach extracts traditional quantitative features along with deep features from one-dimensional convolutional networks as characteristics of prediction samples. The method produces index trend predictions by constructing a graph structure using constituent stock data of indices and applying graph convolution to different index sample features. The proposed methodology's efficacy is verified by utilizing 42 widely employed indicators in the A-share market. The experimental findings demonstrate that when utilizing mean absolute error (MAE) and mean squared error (MSE) as the loss functions for model training, G-Conv outperforms classic methods such as GC-CNN and ADGAT. Specifically, G-Conv reduces the average prediction errors by 5.10% and 4.20% respectively, as evaluated by the two error criteria. Additionally, G-Conv exhibits favorable generalization performance. [ABSTRACT FROM AUTHOR]

Purpose: The biggest obstacle to the sustainable development of online brand communities is social loafing. Based on 3M Hierarchical Model of Personality, this paper aims to discuss the influence of the Big Five on social loafing in online brand communities, and consider the mediating role of dehumanization from the perspective of moral disengagement theory, so as to alleviate social loafing and promote the sustainable development of online brand communities. Design/methodology/approach: The data were collected utilizing the online self-report questionnaires. Community members of mobile phone brands represented by HUAWEI and Xiaomi completed the questionnaire sets. Based on the responses of 214 participants, a structural equation modeling was applied to examine the relationship between Big Five on social loafing in online brand communities. Later, bootstrap analysis was used to further explore the indirect effect of dehumanization. Findings: Only two personality traits, neuroticism and extroversion, have direct effect on social loafing, while agreeableness, conscientiousness and openness not. However, all the Big Five personality traits can indirectly influence social loafing through dehumanization. Originality/value: This study introduces the concept of social loafing into online brand communities, and analyzes social loafing from the unique research perspective of moral disengagement theory, which can not only contribute to the development of moral disengagement theory and social loafing theory but also provide guiding points for relieving social loafing in online brand communities. [ABSTRACT FROM AUTHOR]

Computer-aided diagnosis system is becoming a more and more important tool in clinical treatment, which can provide a verification of the doctors' decisions. In this paper, we proposed a novel abnormal brain detection method for magnetic resonance image. Firstly, a pre-trained AlexNet was modified with batch normalization layers and trained on our brain images. Then, the last several layers were replaced with an extreme learning machine. A searching method was proposed to find the best number of layers to be replaced. Finally, the extreme learning machine was optimized by chaotic bat algorithm to obtain better classification performance. Experiment results based on 5 × hold-out validation revealed that our method achieved state-of-the-art performance. [ABSTRACT FROM AUTHOR]

• Proposal of multifarious methodology with diverse stages for stock market forecast modelling. • The significance of Hurst exponent and Entropy-based Indicators to attain optimal forecasting. • Hurst exponent as a determining and prominent indicator for stock market indices. • Guidance in volatile and uncertain financial markets characterized by unexpected developments. • Characterizing complexity and self-similarity based on Fractal and Entropy analyses. Complex systems constitute components that interact with one another and involve phenomena which are not always easy to understand in terms of their components and interactions. Alternative mathematical models have been developed so that the users' tasks can be facilitated and an actual assistance can be provided for decision-making processes in case of every encountered incident which requires critical decision-making. Within this framework, financial systems can be regarded as complex systems with their volatile and vulnerable nature along with various parameters and interactions involved. Forecasting shifts in stock indices is crucial to validate the potential strategies of monetary mechanisms. Therefore, forecasting is an essential step in financial decision-making to manage data selection and attain robust prediction. Our purpose is to optimize the stock indices ' forecasting model in the stock indices dataset, constructed from the daily values. The following steps were applied to demonstrate the critical significance of Hurst exponent (HE) computed by Rescaled Range (R/S) fractal analysis when used as indicator in conjunction with Shannon entropy (SE) and Renyi entropy (RE) for the future forecasting ability of the stock indices. With this aim, the following stages were performed with indicators obtained from the applications and added into the dataset in the respective order. The first stage consists of: i) HE indicator, ii) Entropy based SE and RE indicators, iii) HE, SE and RE indicators. As the second stage, stock indices day-to-day valuation was evaluated using Multi Layer Regression (MLR), Support Vector Regression (SVR) and Feed Forward Back Propagation (FFBP) algorithms, applied for each indicator for comparative analysis. When compared with earlier works, no relevant work exists in the literature in which the algorithms and above-mentioned indicators have been used in conjunction with one another. This paper, through the multistage methodology and proposed model, demonstrates that HE is obviously a significant and critical determining indicator compared to RE and SE indicators for forecasting purposes. Consequently, experimental results demonstrate the accuracy and applicability of the proposed method. Thus, this study attempts to illustrate a new frontier in domains concerning critical decision-making processes in non-linear, dynamic, and volatile environments. [ABSTRACT FROM AUTHOR]

The competitive segmentation of fuzzy clustering is utilized in a greater manner to deal with the local spatial information of input medical images. Fuzzy clustering favours lesions and tumour identification through the segmentation process where less accuracy attainment and time complexity might be instigated for the identification of oddities. To rectify the above‐said problems, a novel methodology that encapsulates the combination of unsupervised neural network and fuzzy clustering processes, which effortlessly distinguishes the lesion and tumour region in MR brain images is developed through this study. The initial process of the proposed algorithm employs the histogram‐based feature extraction of the input images; whereof, a feature vector selection is made for the operation of self‐organizing map (SOM), which is a neural network functionary that progresses through the mapping process. Modification regarding the membership function of fuzzy entropy clustering (MFEC) is done based on the entropy value of the input image that results in quicker convergence. Finally, the updated objective function of MFEC algorithm augments the SOM result. It is found that the proposed SOM based MFEC algorithm is superior to other traditional segmentation algorithms, which have rendered the better visible understanding of the image. Further, the end‐results of the algorithm are verified through the evaluation of quantity metrics using ground truth of the brain MR images. The proposed SOM based MFEC algorithm precisely provides 82.26% of Jaccard value and 90.05% of Dice Overlap Index value, and these values prove better brain slices segmentation and provide enormous help to radiologists during patient diagnosis. [ABSTRACT FROM AUTHOR]

• Decorrelation formulation based contrast improvement. • Lesion segmentation using modified MASK RCNN. • Transfer Learning based CNN features are extracted. • A Entropy-controlled LS-SVM based best CNN features are selected. Malignant melanoma is considered to be one of the deadliest types of skin cancers which is responsible for the massive number of deaths worldwide. According to the American Cancer Society (ACS), more than a million Americans are living with this melanoma. Since 2019, 192,310 new cases of melanoma are registered, where 95,380 are noninvasive, and 96,480 are invasive. The numbers of deaths due to melanoma in 2019 alone are 7,230, comprising 4,740 men and 2,490 women. Melanoma may be curable if diagnosed at the earlier stages; however, the manual diagnosis is time-consuming and also dependent on the expert dermatologist. In this work, a fully automated computerized aided diagnosis (CAD) system is proposed based on the deep learning framework. In the proposed scheme, the original dermoscopic images are initially pre-processed using the decorrelation formulation technique, which later passes the resultant images to the MASK-RCNN for the lesion segmentation. In this step, the MASK RCNN model is trained using the segmented RGB images generated from the ground truth images of ISBI2016 and ISIC2017 datasets. The resultant segmented images are later passed to the DenseNet deep model for feature extraction. Two different layers, average pool and fully connected, are used for feature extraction, which are later combined, and the resultant vector is forwarded to the feature selection block for down - sampling using proposed entropy-controlled least square SVM (LS-SVM). Three datasets are utilized for validation - ISBI2016, ISBI2017, and HAM10000 to achieve an accuracy of 96.3%, 94.8%, and 88.5% respectively. Further, the performance of MASK-RCNN is also validated on ISBI2016 and ISBI2017 to attain an accuracy of 93.6% and 92.7%. To further increase our confidence in the proposed framework, a fair comparison with other state-of-the-art is also provided. [ABSTRACT FROM AUTHOR]

With the continuous development of the Internet of Vehicles(IoV), Vehicle Edge Computing(VEC) has become a key technology for computational resource scheduling, but more and more smart devices are connected to the internet, which makes it difficult for traditional Vehicle Edge Networks(VEN) to deal with tasks in time. In this paper, in order to cope with the challenges of the large number of devices accessing the internet, we propose a vehicle-assisted computation offloading algorithm based on proximal policy optimization(VCOPPO) for User Equipment(UE) tasks, and it combines dynamic parked vehicles incentives mechanism and computational resource allocation strategy by using road vehicles and parked vehicles as edge servers. Firstly, a non-convex optimization problem combining VEN utility and task processing delay is formulated, subject to the constraints of the residual energy and the transmission rate of the task. Secondly, the proposed VCOPPO is used to solve the formulated non-convex optimization problem, and we use stochastic policy to obtain the optimal computation offloading decisions and resource allocation schemes. Finally, the experimental results have shown that the proposed VCOPPO has an excellent performance in network reward and task processing delay respectively, and it can effectively schedule and allocate computational resources. Compared with using Dueling Deep Q Network(Dueling DQN), Deep Q Network(DQN) and Q-learning methods, the proposed VCOPPO improves the network reward by 31%, 18% and 91%, reduces the delay in task processing by 78%, 63% and 74%, respectively. [ABSTRACT FROM AUTHOR]

• A pre-trained MobileNetV2 was leveraged to extract features from brain MRIs automatically. • The optimal feature layer in MobileNetV2 was obtained by experiment. • Three random neural networks: ELM, RVFL and SNN were trained for classification. • A novel chaotic bat algorithm was employed to improve the performance of classifiers. • The trained model can serve as a visual question answering system and produce accurate results. Computer aided diagnosis systems are playing an important part in clinical treatment. They can help the doctors and physicians to verify the diagnosis decisions. In this study, a new classification algorithm for the brain magnetic resonance image is proposed. Initially, we utilized a MobileNetV2 to extract features from the input brain images, which was pre-trained on ImageNet dataset. Instead of training the deep network, we simply calculate the output of its certain layer to form the feature vector. The optimal feature layer is obtained by the experiment. Then, three different feedforward networks: extreme learning machine, Schmidt neural network and random vector functional-link net, are trained for classification. Chaotic bat algorithm was proposed to optimize the weights and biases in the three randomized neural networks to boost their classification accuracy. The result from 5×hold-out validation reveals that our method can achieve good generalization performance which is comparable to state-of-the-art pathological brain detection methods. The trained model can serve as a visual question answering system and produce accurate results. [ABSTRACT FROM AUTHOR]

In radiomics studies, researchers usually need to develop a supervised machine learning model to map image features onto the clinical conclusion. A classical machine learning pipeline consists of several steps, including normalization, feature selection, and classification. It is often tedious to find an optimal pipeline with appropriate combinations. We designed an open-source software package named FeAture Explorer (FAE). It was programmed with Python and used NumPy, pandas, and scikit-learning modules. FAE can be used to extract image features, preprocess the feature matrix, develop different models automatically, and evaluate them with common clinical statistics. FAE features a user-friendly graphical user interface that can be used by radiologists and researchers to build many different pipelines, and to compare their results visually. To prove the effectiveness of FAE, we developed a candidate model to classify the clinical-significant prostate cancer (CS PCa) and non-CS PCa using the PROSTATEx dataset. We used FAE to try out different combinations of feature selectors and classifiers, compare the area under the receiver operating characteristic curve of different models on the validation dataset, and evaluate the model using independent test data. The final model with the analysis of variance as the feature selector and linear discriminate analysis as the classifier was selected and evaluated conveniently by FAE. The area under the receiver operating characteristic curve on the training, validation, and test dataset achieved results of 0.838, 0.814, and 0.824, respectively. FAE allows researchers to build radiomics models and evaluate them using an independent testing dataset. It also provides easy model comparison and result visualization. We believe FAE can be a convenient tool for radiomics studies and other medical studies involving supervised machine learning. [ABSTRACT FROM AUTHOR]

• First approach in designing Multiple Instance Pooling function through metaheuristics i.e. Genetic algorithm. • A trainable pooling method is proposed - Genetic Pooling. • Genetic Pooling outperforms other MIL pooling methods in terms of AUC and average precision score. Multiple Instance Learning is a weakly supervised learning technique which is particularly well suited for medical data analysis as the class labels are often not available at desired granularity. Multiple Instance Learning through Deep Neural Networks is relatively a new paradigm in machine learning. The most important part of Multiple Instance Learning through Deep Neural Networks is designing a trainable pooling function which determines the instance-to bag relationship. In this paper, we propose a Multiple Instance pooling technique based on Genetic Algorithm called Genetic Pooling. In this technique, instance labels inside a bag are optimized by minimizing bag-level losses. The main contribution of the paper is that the bag level pooling layer for generating attention weights for bag instances are replaced by random initialization of attention weights and finding the optimized attention weights through Genetic Algorithm. [ABSTRACT FROM AUTHOR]

Background: To investigate the predictive ability of high-throughput MRI with deep survival networks for biochemical recurrence (BCR) of prostate cancer (PCa) after prostatectomy. Methods: Clinical-MRI and histopathologic data of 579 (train/test, 463/116) PCa patients were retrospectively collected. The deep survival network (iBCR-Net) is based on stepwise processing operations, which first built an MRI radiomics signature (RadS) for BCR, and predicted the T3 stage and lymph node metastasis (LN+) of tumour using two predefined AI models. Subsequently, clinical, imaging and histopathological variables were integrated into iBCR-Net for BCR prediction. Results: RadS, derived from 2554 MRI features, was identified as an independent predictor of BCR. Two predefined AI models achieved an accuracy of 82.6% and 78.4% in staging T3 and LN+. The iBCR-Net, when expressed as a presurgical model by integrating RadS, AI-diagnosed T3 stage and PSA, can match a state-of-the-art histopathological model (C-index, 0.81 to 0.83 vs 0.79 to 0.81, p > 0.05); and has maximally 5.16-fold, 12.8-fold, and 2.09-fold (p < 0.05) benefit to conventional D'Amico score, the Cancer of the Prostate Risk Assessment (CAPRA) score and the CAPRA Postsurgical score. Conclusions: AI-aided iBCR-Net using high-throughput MRI can predict PCa BCR accurately and thus may provide an alternative to the conventional method for PCa risk stratification. [ABSTRACT FROM AUTHOR]