Due to the implementation and performance limitations of centralized learning automatic modulation classification (CentAMC) method, this paper proposes a decentralized learning AMC (DecentAMC) method using model consolidation and lightweight design. Specifically, the model consolidation is realized by a central device (CD) for edge device (ED) model averaging (MA) and multiple EDs for ED model training. The lightweight is designed by separable convolutional neural network (S-CNN), in which the separable convolutional layer is utilized to replace the standard convolution layer and most of fully connected layers are cut off. Simulation results show that the proposed method substantially reduces the storage and computational capacity requirements of the EDs and communication overhead. The training efficiency also shows remarkable improvement. Compared with convolutional neural network (CNN), the space complexity (i.e., model parameters and output feature map) is decreased by about 94% and the time complexity (i.e., floating point operations) of S-CNN is decreased by about 96% while degrading the average correct classification probability by less than 1%. Compared with S-CNN-based CentAMC, without considering model weights uploading and downloading, the training efficiency of our proposed method is about <inline-formula> <tex-math notation="LaTeX">${N}$ </tex-math></inline-formula> times of it, where <inline-formula> <tex-math notation="LaTeX">${N}$ </tex-math></inline-formula> is the number of EDs. Considering the model weights uploading and downloading, the training efficiency of our proposed method can still be maintained at a high level (e.g., when the number of EDs is 12, the training efficency of the proposed AMC method is about 4 times that of S-CNN-based CentAMC in dataset <inline-formula> <tex-math notation="LaTeX">$D_{1} = \{2{\mathrm {FSK, 4FSK, 8FSK, BPSK, QPSK, 8PSK, 16QAM}}\}$ </tex-math></inline-formula> and about 5 times that of S-CNN-based CentAMC in dataset <inline-formula> <tex-math notation="LaTeX">$D_{2} = \{2 {\mathrm {FSK, 4FSK, 8FSK, BPSK, QPSK, 8PSK, PAM2, PAM4, PAM8, 16QAM}}\}$ </tex-math></inline-formula>), while the communication overhead is reduced more than 35%.

Automatic modulation classification (AMC) is a promising technology for identifying modulation types, and deep learning (DL)-based AMC is one of its main research directions. Conventional DL-based AMC methods are centralized solutions (i.e., CentAMC), which are trained on abundant data collected from local clients and stored in the server and generally have advanced performance, but their major problem is the risk of data leakage. Besides, if DL-based AMC is only trained with the data from their corresponding clients, it may exhibit weak performance. Thus, a federated learning (FL)-based AMC (FedeAMC) is proposed under the condition of class imbalance and noise varying. Its advantage is low risk of data leakage without severe performance loss, because data and training are in each local client, while only knowledge (i.e., gradient or model weight), rather than data, is shared with the server. In addition, there is generally class imbalance problem in each local client, and balanced cross entropy is introduced as loss function for solving this problem. Simulation results demonstrated that average accuracy gap between FedeAMC and CentAMC is less than 2%.

Radio frequency fingerprint (RFF) identification is a popular topic in the field of physical layer security. However, machine learning based RFF identification methods require complicated feature extraction manually while deep learning based methods are hard to achieve robust identification performance. To solve these problems, we propose a novel RFF identification method based on heat constellation trace figure (HCTF) and slice integration cooperation (SIC). HCTF is utilized to avoid the manual feature extraction and SIC is adopted to extract more features automatically in RF signals. Experimental results show that our proposed HCTF-SIC identification method can achieve higher accuracy than the existing RFF methods. The identification accuracy achieves 91.07% when SNR <inline-formula> <tex-math notation="LaTeX">$\pmb {=}$ </tex-math></inline-formula> 0 dB and it is even higher than 99.64% when the SNR <inline-formula> <tex-math notation="LaTeX">$\pmb {\ge }$ </tex-math></inline-formula> 5 dB.

Background: Arrhythmias are common problems in hypertensive patients. The presence and complexity of both supraventricular and ventricular arrhythmias may influence morbidity, mortality, as well as the quality of life of patients. Objective: The aim of this study was to assess the diagnostic value of combined 24h BP and ECG Holter monitoring in detection of cardiac arrhythmias in patients with arterial hypertension. Methods: We analyzed the simultaneous records of combined 24h BP and ECG Holter monitoring for 356 adult patients with diagnosed arterial hypertension in the period from January 2017 until January 2021 year. The cardiac arrhythmias were classified in three main groups as following: a) Supraventricular arrhythmias; b) Ventricular arrhythmias; c) Bradyarrhythmia’s. Standard transthoracic echocardiograms were performed in order to evaluate signs of hypertensive or structural heart disease with focus on left ventricle hypertrophy and LV function. Results: Patients had a mean age of 64 ± 11years, 62% male. Average clinic BP was 153.4+18/87.5+14 mmHg. More than 46% of patients displayed a very high-risk profile. In all enrolled patients, cardiac arrhythmia was detected in 302 (84%) patients. The total number of patients with supraventricular arrhythmias was 153 (50,7%). Ventricular arrhythmias were detected in 98 (32,5%) patients. Bradyarrhythmia’s were detected in 51 (16,9%) patients. Elevated resting heart rate in sinus rhythm was detected in 87 (31,6%) of 275 patients with sinus rhythm. Conclusion: Most arrhythmias are related to longstanding arterial hypertension. Effective treatment of arterial hypertension plays important role in preventing structural and functional cardiac abnormalities which will contribute to the reduction of cardiac arrhythmias in hypertensive patients.

BACKGROUND The pre-analytical phase, which includes all preparatory actions to the analytical procedure, is part of the process during which there is the greatest possibility of laboratory errors. This study was conducted to investigate the frequency and types of laboratory errors during work in the clinical laboratory as well as the frequency and types of laboratory errors in the pre-analytical phase of laboratory work. METHODS The retrospective, descriptive study covered the period from 01/01/2016 to 12/31/2016 within which the presence of 5 different indicators of quality of work, i.e., pre-analytical errors, was monitored: improperly drawn blood, coagulated blood sample, hemolyzed blood sample, improperly marked referral for analysis, and insufficient sample for analysis. RESULTS The most common error in the pre-analytical phase of our study was "coagulated sample", followed by: "improperly drawn blood", "improperly marked referral", "insufficient sample for analysis", and "hemolyzed sample". Using the chi-squared test, a statistically significant difference was found in the frequency of occurrence of certain types of indicators in different departments (p < 0.005). CONCLUSIONS Reduction of these errors can be achieved through analyzing and correcting the reasons for them, education, and by joint action of experts and international organizations, continual training of staff as well as to following the adopted guidelines and standards.

Architectural patterns are solutions to common problems in software design. These usually tackle one or more customer requirements (business, functional, or quality requirements). As continuous deployment becomes more important for satisfying customer’s experience in software projects, there is a greater need in supporting continuous integration as its prerequisite. The nature of architectural design patterns is that some are more suitable for continuous integration then others. However, the existing approaches do not prescribe how to tailor architectural design considering integration challenges.To identify patterns suitable for continuous integration, we present a methodology for identification of architectural design patterns that support continuous integration based on continuous assessment. Our methodology, based on Aglie, enables DevOps engineers to assess their integration experience considering design changes, implementation changes, and creation of new version deliverables. The methodology complements common DevOps activities and introduces templates for communicating feedback from DevOps engineers to architects. Architects handle the communicated feedback as requirements and optimise their design accordingly. In this way, the design decisions take shape of solution patterns that, besides other existing requirements, satisfy requirements necessary to facilitate continuous integration.

Highlights • Gap of public on-the-ground cancer data was addressed through questionnaires to professionals in the SEE region.• There is a lack of diagnostic imaging and radiotherapy modalities in the SEE region in comparison to Western Europe.• The Mortality to incidence ratio correlates inversely with the economic development and the availability of radiotherapy equipment.• The cancer incidence in SEE countries was correlate directly with the life expectancy and the availability of diagnostic equipment.• The need for reliable national and regional cancer registries in SEE countries for data collection and analysis has been emphasized.

We investigated DAB1-protein deficiency in the inner-ear development of yotari in comparison to humans and wild-type (wt) mice by immunofluorescence for the expression of connexins (Cxs) and the pannexin Panx1. The spatial and temporal dynamics of Cx26, Cx32, Cx37, Cx40, Cx43, Cx45, and Panx1 were determined in the sixth and eighth weeks of human development and at the corresponding mouse embryonic E13.5 and E15.5, in order to examine gap junction intercellular communication (GJIC) and hemichannel formation. The quantification of the area percentage covered by positive signal was performed for the epithelium and mesenchyme of the cochlear and semicircular ducts and is expressed as the mean ± SD. The data were analysed by one-way ANOVA. Almost all of the examined Cxs were significantly decreased in the cochlear and semicircular ducts of yotari compared to wt and humans, except for Cx32, which was significantly higher in yotari. Cx40 dominated in human inner-ear development, while yotari and wt had decreased expression. The Panx1 expression in yotari was significantly lower than that in the wt and human inner ear, except at E13.5 in the mesenchyme of the wt and epithelium and mesenchyme of humans. Our results emphasize the relevance of GJIC during the development of vestibular and cochlear functions, where they can serve as potential therapeutic targets in inner-ear impairments.

Recently, much attention has been paid to the reuse of bauxite residues from alumina production, also known as red mud, in the cement industry. Red mud bears the potential to improve concrete properties due to its favourable chemical composition and particle size. In this work, the synergy between locally available red mud and common supplementary cementitious materials such as fly ash, slag, calcined clay and limestone in cement mixes is investigated. All materials used were sourced from the immediate vicinity of the cement plant. The study of synergy involved the evaluation of the individual chemical reactivity of each material using the R3 test by isothermal calorimetry as well as their joint contribution to the heat of hydration and the composition of the reaction products of the paste and the compressive strength of the mortar. The results show how, by understanding the synergy between the materials, a higher level of cement substitutions can be achieved without compromising the mechanical properties of the mortar.

The majority of people with Autism Spectrum Disorder (ASD) exhibit difficulties in social communication and behavior, which hinder their learning capability, amid others. Among technological solutions for people with ASD, serious games are frequently used to enhance learning of specific skills and instructional contents. However, because of heterogeneity in applications and game design, few studies have investigated their use in training daily activities. This paper presents a 3D personalized serious game we developed and validated to help ASD patients practice with shopping activities. Personalized training is paramount in people with ASD, thus several elements of this game were personalized to improve engagement and therefore the effectiveness of the virtual training. In order to assess the validity of the game, ten subjects (age $\text{11.9} \pm \text{2.7}$, 20% female) with ASD played ten sessions of the serious game, once per week. The participants underwent a real-life experience pre- and post-training in a real-life supermarket. Changes in daily living skills among participants were evaluated through specific tools: a form based on the International Classification of Functioning, Disability and Health for Children and Youth; and the Vineland Adaptive Behavior Scale II. Significant improvements (p$< $0.05) were detected in the main skills trained with the serious game, especially in learning the shopping procedure, directing attention, and problem-solving skills. These findings suggest that personalized serious games can represent a prominent tool to enhance daily living skills, but future work should clinically validate their efficacy.