Secure machine learning against adversarial samples at test time

Deep neural networks (DNNs) are widely used to handle many difficult tasks, such as image classification and malware detection, and achieve outstanding performance. However, recent studies on adversarial examples, which have maliciously undetectable perturbations added to their original samples that are indistinguishable by human eyes but mislead the machine learning approaches, show that machine learning models are vulnerable to security attacks. Though various adversarial retraining techniques have been developed in the past few years, none of them is scalable. In this paper, we propose a new iterative adversarial retraining approach to robustify the model and to reduce the effectiveness of adversarial inputs on DNN models. The proposed method retrains the model with both Gaussian noise augmentation and adversarial generation techniques for better generalization. Furthermore, the ensemble model is utilized during the testing phase in order to increase the robust test accuracy. The results from our extensive experiments demonstrate that the proposed approach increases the robustness of the DNN model against various adversarial attacks, specifically, fast gradient sign attack, Carlini and Wagner (C&W) attack, Projected Gradient Descent (PGD) attack, and DeepFool attack. To be precise, the robust classifier obtained by our proposed approach can maintain a performance accuracy of 99% on average on the standard test set. Moreover, we empirically evaluate the runtime of two of the most effective adversarial attacks, i.e., C&W attack and BIM attack, to find that the C&W attack can utilize GPU for faster adversarial example generation than the BIM attack can. For this reason, we further develop a parallel implementation of the proposed approach. This parallel implementation makes the proposed approach scalable for large datasets and complex models.

The Effect of Sanitizing Treatments on Respirator Filtration Performance

The potential for alcoholic vapors emitted by common sanitizing treatments to deteriorate the (electrostatic) filtration performance of disposable respirator masks has been investigated. Reports in the literature and some standard test methods provide a confusing and ambiguous picture concerning the relevance of this effect. Four different types of exposure were investigated in this study to clarify the effect of alcoholic vapor emissions on respirator masks. These included exposure to saturated vapors, use of hand sanitizers, cleaning of table surfaces and sanitization of masks by spraying them with alcohol-containing solutions. Methods employed were designed to be as real-world oriented as possible while remaining reproducible. Filtration performance and deterioration effects on exposure to the different treatments were determined on three different types of certified commercial respirator masks—a P2 and two KN95 masks. This study provides substantial evidence that disposable respirator masks with an accepted performance rating are seriously compromised from an exposure to saturated alcoholic vapors, can tolerate a one-off spray treatment with an alcoholic solution and retain their attested protection under the influence of alcoholic vapors from the use of hand sanitizer or spray sanitizer. Considering the range of vastly different outcomes obtained from the four treatments investigated, it seems prudent to assess in each case the specific effects of alcoholic solution treatments and vapors on respirator masks before use.

The 2020 “Padua Criteria” for Diagnosis and Phenotype Characterization of Arrhythmogenic Cardiomyopathy in Clinical Practice

Arrhythmogenic Cardiomyopathy (ACM) is a heredo-familial cardiac disease characterized by fibro-fatty myocardial replacement and increased risk of sudden cardiac death. The diagnosis of ACM can be challenging due to the lack of a single gold-standard test: for this reason, it is required to satisfy a combination of multiple criteria from different categories including ventricular morpho-functional abnormalities, repolarization and depolarization ECG changes, ventricular arrhythmias, tissue characterization findings and positive family history/molecular genetics. The first diagnostic criteria were published by an International Task Force (ITF) of experts in 1994 and revised in 2010 with the aim to increase sensitivity for early diagnosis. Limitations of the 2010 ITF criteria include the absence of specific criteria for left ventricle (LV) involvement and the limited role of cardiac magnetic resonance (CMR) as the use of the late gadolinium enhancement technique for tissue characterization was not considered. In 2020, new diagnostic criteria (“the Padua criteria”) were proposed. The traditional organization in six categories of major/minor criteria was maintained. The criteria for identifying the right ventricular involvement were modified and a specific set of criteria for identifying LV involvement was created. Depending on the combination of criteria for right and LV involvement, a diagnosis of classic (right dominant) ACM, biventricular ACM or left-dominant ACM is then made. The article reviews the rationale of the Padua criteria, summarizes the main modifications compared to the previous 2010 ITF criteria and provides three examples of the application of the Padua criteria in clinical practice.

Process optimization for the production of biodiesel from Azolla Microphylla oil and its fuel characterization

The most competent and operative use of renewable feedstock is super critical for the production of biodiesel which has increased attention worldwide pertaining to aquatic fern Azolla. Maximizing the biodiesel yield by optimizing the process parameters of the low-frequency ultrasonic energy-assisted transesterification process of Azolla oil is the need of the hour for minimizing the production cost of biodiesel. Response Surface Methodology (RSM) was applied using central composite rotatable design (CCRD) to find the best optimum reaction parameters for this transesterification process. The optimized reaction parameters arrived from the design of experiments were as following: methanol/Azolla oils molar ratio (A) = 6.49 mole/mole, KOH catalyst concentration (B) = 1.69 (weight% of oil), reactiion time (C) = 34.74 min and reaction temperature (D) = 38.87°C. The best higher theoretical predicted Azolla Fatty Acid Methyl Ester (FAME) yield was Y = 99.76% which is in well coincidence with the actual yield. The extracted Azolla biodiesel was tested for various fuel properties with standard test procedures and found to be in agreement with various Biodiesel standards and the results are promising in terms of utilizing Azolla oil as an inexhaustible and potentially economical source of biodiesel.

Identification of asymptomatic Leishmania infections: a scoping review

Background Asymptomatic Leishmania infection may play an important role in the transmission of the parasite in endemic areas. At present there is no consensus on the definition of asymptomatic Leishmania infection, nor is there a safe and accessible gold standard test for its identification. Methods This paper presents a scoping review to summarize definitions of asymptomatic Leishmania infection found in the literature, as well as to detail the approach (molecular, serological, cellular, and/or parasitological tests) used by researchers to identify this asymptomatic population. A scoping review of published and gray literature related to asymptomatic Leishmania infection was conducted; retrieved citations were screened based on predefined eligibility criteria, and relevant data items were extracted from eligible articles. The analysis is descriptive and is presented using tables, figures, and thematic narrative synthesis. Results We conducted a screening of 3008 articles, of which 175 were selected for the full review. Of these articles, we selected 106 that met the inclusion criteria. These articles were published between 1991 and 2021, and in the last 5 years, up to 38 articles were reported. Most of the studies were conducted in Brazil (26%), Spain (14%), India (12%), Bangladesh (10%), and Ethiopia (7%). Of the studies, 84.9% were conducted in the immunocompetent population, while 15.1% were conducted in the immunosuppressed population (HIV, immunosuppressive drugs, and organ transplantation population). We report 14 different techniques and 10 strategies employed by researchers to define asymptomatic Leishmania infection in an endemic area. Conclusions The definition of asymptomatic Leishmania infection is not unified across the literature, but often includes the following criteria: residence (or extended stay) in a Leishmania-endemic area, no reported signs/symptoms compatible with leishmaniasis, and positive on a combination of serological, molecular, cellular, and/or parasitological tests. Caution is recommended when comparing results of different studies on the subject of asymptomatic infections, as the reported prevalence cannot be confidently compared between areas due to the wide variety of tests employed by research groups. More research on the importance of asymptomatic immunosuppressed and immunocompetent Leishmania-positive populations in leishmaniasis epidemiology is required. Graphical Abstract

ELECTRICAL PARAMETERS ESTIMATION OF SINGLE DIODE PV MODULE MODEL USING HYBRID METAHEURISTIC ALGORIHM

In this paper, an improved hybrid algorithm called differential evolution with integrated mutation per iteration (DEIM) is proposed to extract five parameters of single-diode PV module model obtained by combining differential evolution (DE) algorithm and electromagnetic-like (EML) algorithm. The EML algorithm's attraction-repulsion idea is employed in DEIM in order to enhance the mutation process of DE. The proposed algorithm is validated with other methods using experimental I-V data. The results of presented method reveal that simulated I-V characteristics have a high degree of agreement with experimental ones. The proposed model has an average root mean square error of 0.062A, an absolute error of 0.0452A, a mean bias error of 0.006A, a coefficient of determination of 0.992, a standard test deviation around 0.04540, and 15.33sec as execution time. The results demonstrate that the proposed method is better in terms of the accuracy and execution time (convergence) when compared with other methods where provide less errors.

The Influencing Factors for Volume Stability of Ladle Slag

The purpose of this study was to investigate the mechanism causing the unsoundness of ladle slag. Calcination temperature may have an impact on the level of reactivity of f-CaO. When CaO was produced at a higher temperature, the reactivity of CaO was lower. For example, dead burnt CaO (DCaO) was produced at higher temperatures than light burnt CaO (LCaO); therefore, DCaO had less reactivity than LCaO. In a hydration test, DCaO (1500 °C) showed 62 times lower reactivity than LCaO (900 °C), which meant that DCaO would result in the delay of hydration of CaO easily. Additionally, DCaO would cause unsoundness more easily than LCaO when adding the same number of cementitious materials. For this reason, using ASTM C114-18 (Standard Test Methods for Chemical Analysis of Hydraulic Cement) to quantify DCaO content may underestimate DCaO content by up to 20%. Conversely, this method was more suitable for f-CaO since it had high reactivity. Moreover, this study demonstrated that ladle slag would cause unsoundness when added into the cementitious material because it was produced from a higher temperature process (over 1500 °C), which generates the DCaO. Therefore, when reusing ladle slag, the problem of low reactivity of DCaO should be considered.

Proposed Power Systems Planning in Indian Scenario for Integrating EV Charging Infrastructure

The growing concern about fossil energy exhaustion, air pollution, and ecological deprivation has made electric vehicles (EVs) a practical option in contrast to combustion engine-driven vehicles. In any case, driving extent uneasiness is one of the innate inadequacies related with EVs. Massive integration of EV charging load into the power system may be a threat to the distribution network. Spontaneous situation of charging stations in the distribution system and uncoordinated charging will augment the load demand thereby resulting in voltage instability, deterioration of reliability indices, harmonic distortions, and escalated power losses. This chapter will concentrate on breaking down the effect of EV chargers on the working parameters, for example, voltage dependability, unwavering quality, and force misfortune. The examination will be completed on standard test systems. The discoveries of the proposed part will evaluate the effect of EV charging load on the working parameters of the distribution system and help in proposing a framework for charging station planning.