Search

Given the rising challenge of client-based web attacks through vulnerabilities in websites, traditional pattern detection methods often fall short in identifying emerging threats. To bridge this gap, our study proposes a methodology employing machine learning algorithms to counteract three specific types of client-based web attacks: Malicious JavaScript, phishing attacks, and script-based web attacks. Our method extracts significant features from the source code and URLs, subsequently applying a range of machine learning models, including random forest (RF), deep neural network (DNN), and convolutional neural network (CNN), to pinpoint the most effective model. Experimental evidence from our research highlights the RF model’s exceptional accuracy, achieving 99.99% in detecting Malicious JavaScript, 95.11% for phishing attacks, and 94.77% for script-based web attacks. Additionally, our work extends beyond theoretical contributions, evidenced by the development of a Chrome extension based on the high-performing RF model, offering a tangible solution for enhancing web browsing security. [ABSTRACT FROM AUTHOR]

Malalignment in the lower limb structure occurs due to various causes. Accurately evaluating limb alignment in situations where malalignment needs correction is necessary. To create an automated support system to evaluate lower limb alignment by quantifying mechanical tibiofemoral angle (mTFA), mechanical lateral distal femoral angle (mLDFA), medial proximal tibial angle (MPTA), and joint line convergence angle (JLCA) on full-length weight-bearing radiographs of both lower extremities. In this retrospective study, we analysed 404 radiographs from one hospital for algorithm development and testing and 30 radiographs from another hospital for external validation. The performance of segmentation algorithm was compared to that of manual segmentation using the dice similarity coefficient (DSC). The agreement of alignment parameters was assessed using the intraclass correlation coefficient (ICC) for internal and external validation. The time taken to load the data and measure the four alignment parameters was recorded. The segmentation algorithm demonstrated excellent agreement with human-annotated segmentation for all anatomical regions (average similarity: 89–97%). Internal validation yielded good to very good agreement for all the alignment parameters (ICC ranges: 0.7213–0.9865). Interobserver correlations between manual and automatic measurements in external validation were good to very good (ICC scores: 0.7126–0.9695). The computer-aided measurement was 3.44 times faster than was the manual measurement. Our deep learning-based automated measurement algorithm accurately quantified lower limb alignment from radiographs and was faster than manual measurement. [ABSTRACT FROM AUTHOR]

Tests of visual search can index the effects of perceptual load and compare the processing efficiency for different object types, particularly when one examines the set-size effect, the increase in search time for each additional stimulus in an array. Previous studies have shown that the set-size effect is increased by manoeuvres that impede object processing, and in patients with object processing impairments. In this study, we examine how the low-level visual impairment of hemianopia affects visual search for complex objects, using a virtual paradigm. Forty-two healthy subjects performed visual search for faces, words, or cars with full-viewing as well as gaze-contingent simulations of complete left or right hemianopia. Simulated hemianopia lowered accuracy and discriminative power and increased response times and set-size effects, similarly for faces, words and cars. A comparison of set-size effects between target absent and target present trials did not show a difference between full-view and simulated hemianopic conditions, and a model of decision-making suggested that simulated hemianopia reduced the rate of accumulation of perceptual data, but did not change decision thresholds. We conclude that simulated hemianopia reduces the efficiency of visual search for complex objects, and that such impairment should be considered when interpreting results from high-level object processing deficits. [ABSTRACT FROM AUTHOR]

Angle measurement methods for measuring pes planus may lose consistency by errors between observers. If the feature points for angle measurement can be provided in advance with the algorithm developed through the deep learning method, it is thought that the error between the observers can be reduced. A total of 300 weightbearing lateral radiographs were used for the development of the deep learning-based algorithm, and a total of 95 radiographs were collected for the clinical validation test set. Meary angle (MA) and calcaneal pitch (CP) were selected as measurement methods and measured twice by three less-experienced physicians with the algorithm-based tool and twice without. The intra- and inter-observer agreements of MA and CP measures were assessed via intra-class correlation coefficient. In addition, verification of the improvement of measurement performance by the algorithm was performed. Interobserver agreements for MA and CP measurements with algorithm were more improved than without algorithm. As for agreement with reference standard, combining the results of all readers, both MA and CP with algorithm were greater than those without algorithm. The deep learning algorithm tool is expected to improve the reproducibility of radiographic measurements for pes planus, especially by improving inter-observer agreement. [ABSTRACT FROM AUTHOR]

The ocean contains numerous marine organisms, including algae, animals, and plants, from which diverse marine polysaccharides with useful physicochemical and biological properties can be extracted. In particular, fucoidan, carrageenan, alginate, and chitosan have been extensively investigated in pharmaceutical and biomedical fields owing to their desirable characteristics, such as biocompatibility, biodegradability, and bioactivity. Various therapeutic efficacies of marine polysaccharides have been elucidated, including the inhibition of cancer, inflammation, and viral infection. The therapeutic activities of these polysaccharides have been demonstrated in various settings, from in vitro laboratory-scale experiments to clinical trials. In addition, marine polysaccharides have been exploited for tissue engineering, the immobilization of biomolecules, and stent coating. Their ability to detect and respond to external stimuli, such as pH, temperature, and electric fields, has enabled their use in the design of novel drug delivery systems. Thus, along with the promising characteristics of marine polysaccharides, this review will comprehensively detail their various therapeutic, biomedical, and miscellaneous applications. [ABSTRACT FROM AUTHOR]

This study evaluated the moisturizing effect of serine-loaded solid lipid nanoparticles (serine-SLN) and polysaccharide-rich reed ( Phragmites communis) root extract (RRE) incorporated in hydrogel bases. The hydrogels with serine-SLN and/or RRE were carefully applied on the volar forearm of human volunteers. Their moisturizing efficacy was evaluated by monitoring conductance values using a skin surface hygrometer. The values of the area under the normalized conductance-time curve (AUCC) were developed and compared as a parameter for the water holding capacity of the skin. Hydrogels with serine-SLN did not significantly moisturize the skin, while hydrogel containing 0.25% RRE produced a significant increase in the moisture content of the skin. However, adding more than 0.25% of RRE into the hydrogel base decreased the moisturizing effect due to the marked reduction of viscosity. Significantly enhanced moisturizing effect was observed with the hydrogel containing 0.25% RRE and 3% serine-SLN, with AUCC increased 2.21 times compared to than blank hydrogel. The results imply that effective delivery of serine into the skin is possible using lipid-based nanocarriers and RRE, which could be a promising strategy to moisturize the skin effectively. [ABSTRACT FROM AUTHOR]

We evaluated the feasibility of using quantitatively measured thoracic components, as compared to body mass index (BMI), for predicting the image noise of coronary computed tomography angiography (CCTA). One hundred subjects (M:F = 64:36; mean age, 55 ± 8.8 years) who underwent prospective electrocardiography-gated CCTA and low-dose chest computed tomography (CT) were analyzed retrospectively. The image noise of the CCTA was determined by the standard deviation of the attenuation value in a region of interest on the aortic root level. On the low-dose chest CT, the areas of the thoracic components were measured at the aortic root level. An auto-segmentation technique with the following threshold levels was used: quantitatively measured area of total thorax [QMAtotal: -910 to 1000 Hounsfield units (HU)], lung (QMAlung: -910 to -200 HU), fat (QMAfat: -200 to 0 HU), muscle (QMAmuscle: 0-300 HU), soft tissue (fat + muscle, QMAsoft tissue: -200 to 300 HU), bone (QMAbone: 300-1000 HU) and solid tissue (fat + muscle + bone, QMAsolid tissue: -200 to 1000 HU). The relationship between image noise and variable biometric parameters including QMA was analyzed, and the linear correlation coefficients were used as indicators of the strength of association. Among the variable biometric parameters, including BMI, QMAsolid tissue showed the highest correlation coefficient with image noise in all subjects (r = 0.804), males (r = 0.716), females (r = 0.889), the overweight (r = 0.556), and the non-overweight subgroups (r = 0.783). QMAsolid tissue can be used as a potential surrogate predictor of the image noise level in low tube voltage CCTA. [ABSTRACT FROM AUTHOR]

Correction to: I Scientific Reports i https://doi.org/10.1038/s41598-022-16995-6, published online 28 July 2022 In the original version of this Article Jalim Koo and Sangchul Hwang were omitted as equally contributing authors. The original article can be found online at https://doi.org/10.1038/s41598-022-16995-6. [Extracted from the article]