Additional Modules Research Articles

Distillation embedded absorbable pruning for fast object re-identification

Combining knowledge distillation (KD) and network pruning (NP) shows promise in learning a light network to accelerate object re-identification. However, KD requires an untrained student network to establish more critical connections in early epochs, but NP demands a well-trained student network to avoid destroying critical connections. This presents a dilemma, potentially leading to a collapse of the student network and harming object Re-ID performance. For that, we propose a distillation embedded absorbable pruning (DEAP) method. We design a pruner-convolution-pruner (PCP) unit to resolve the dilemma by loading NP’s sparse regularization on extra untrained pruners. Additionally, we propose an asymmetric relation knowledge distillation method. It readily transfers feature representation knowledge and asymmetric pairwise similarity knowledge without using additional adaptation modules. Finally, we apply re-parameterization to absorb pruners of PCP units to simplify student networks. Experiments demonstrate the superiority of DEAP, such as on the VeRi-776 dataset, with ResNet-101 as a teacher, DEAP saves 73.24% of model parameters and 71.98% of floating-point operations without sacrificing accuracy.

Ultrafast, self-powered monolithic pressure sensing technology induced by piezo-pyrophototronics

Pressure sensing is a pivotal aspect of modern technology that has been extensively explored, but still presents significant challenges. This work introduces an innovative ultrafast self-powered pressure sensing technology by harnessing the lateral photovoltaic effect (LPE) in a CIGS heterojunction via piezo-pyrophototronics. By utilizing non-uniform irradiation, a surface gradient of charge carriers is induced, resulting in the generation of a lateral photovoltage (LPV) that contributes to the excellent self-powered characteristic of the sensor. This gradient is then coupled with external pressure through the piezo-phototronic effect, enabling the sensing capability of the device. The enhancement of LPV, which is dependent on the irradiation position, power and wavelength, exhibits an extremely linear relationship with the applied external pressure ranging from 0 to 2.0 MPa. Besides, the pyroelectric field generated by transient temperature variations introduces an additional modulation to the pressure sensing effect, leading to an impressive pressure sensitivity of 64.25 mV/MPa, and a rapid response speed of 7.8/8.9 μs. This work highlights the benefits of the lateral photovoltaic effect, piezo-phototronic effect, and pyroelectric modulation in developing highly sensitive and ultrafast self-powered pressure sensors.

An Angle-Based Virtual Vector Model Predictive Current Control for IPMSM Considering Overmodulation

The conventional model predictive control method suffers from large current and torque harmonics due to the limited voltage vectors. To increase the control performance, in many papers, the methods of extending the predictive control set by adding virtual vectors are proposed. However, constructing virtual vectors will bring extra computational burden for vector optimization. To solve this problem, this paper proposes an angle-based virtual vector model predictive current control (ABVV-MPCC) method for interior permanent magnet synchronous motor (IPMSM). In this method, a hexagonal coordinate system is introduced to define the virtual vectors. After obtaining the composite relationship between the reference voltage vector and the adjacent basic vector, the vector optimization range is reduced from six voltage sectors to one voltage sector. And the optimal voltage vector is directly obtained by searching for the closest candidate voltage vector. After adopting the proposed method, the exhaustive optimization is avoided and the calculation burden is independent from the virtual voltage vectors number. Therefore, virtual voltage vectors number can be set arbitrarily, which greatly reduces current harmonics. Besides, a simple overmodulation dealing method by finding the optimal reachable voltage vector is proposed. And the dynamic performance under overmodulation is increased without adding additional control module.

The INGUIDE International Guideline Training and Certification Programme

AbstractHealth guidelines impact clinical, public health and policy practice, but there is no regulation for their development, often leading to variability in quality and trustworthiness. The International Guideline Training and Certification Programme (INGUIDE), established by faculty at McMaster University in partnership and under the auspice of the Guidelines International Network (GIN), addresses this shortcoming by offering structured, evidence‐based training and certification for those involved in guidelines (inguide.org). This commentary describes INGUIDE's background, purpose, structure and significance after approximately 3 years of operation. INGUIDE's mission is to enhance the quality of health guidelines globally. It provides comprehensive training to ensure the systematic development of guidelines based on the best evidence and adherence to international quality standards, as reflected in its ISO:9001 certification. The programme also emphasizes capacity building, filling educational gaps, and ensuring the global inclusivity of its courses. INGUIDE's certification covers the entire lifecycle of guideline development and includes several levels of certification, ranging from panel member to methodologist, lead methodologist, developer, chair and instructor certification. The programme already has had a global impact, training over 1500 learners since its launch. INGUIDE is led by a steering committee with input from an international advisory board and operational staff, supported by certified instructors. The programme's vision for the future includes expanding accessibility and creating additional training modules, with a commitment to continuous improvement and adaptation to diverse healthcare contexts, in particular low‐ and middle‐income countries and settings.

Collecting non-native English speech through a web-and mobile-based application

The motivation behind this study is to collect non-native English speech recordings seamlessly. Even though numerous corpora containing non-native English speakers exist, there are not so many resources of Polish speakers uttering sentences in English. That is why a web-based application is built to record L2 English speakers. The same application is also developed in the mobile form as it provides easy access to the software created and, at the same time, built-in microphones ensure sufficient quality of recordings. A set of sentences is prepared, including four main types, i.e., declarative, imperative, interrogative, and exclamatory. Recordings are saved in WAV audio format. Local data are first saved in the memory of the device on which the application is installed, and then, the user may listen to the recordings and then send them to the server using FTP (File Transfer Protocol). The user interface allows for English language level selection, by choosing one of the four possible options, i.e., basic, intermediate, advanced, and fluent. A group of volunteers provided speech samples, resulting in the collection of utterances from more than 600 speakers, available in the created corpus. Future development envisions adding additional modules that will allow for speech signal analyses.

Integrated QSAR Models for Prediction of Serotonergic Activity: Machine Learning Unveiling Activity and Selectivity Patterns of Molecular Descriptors.

Understanding the features of compounds that determine their high serotonergic activity and selectivity for specific receptor subtypes represents a pivotal challenge in drug discovery, directly impacting the ability to minimize adverse events while maximizing therapeutic efficacy. Up to now, this process has been a puzzle and limited to a few serotonergic targets. One approach represented in the literature focuses on receptor structure whereas in this study, we followed another strategy by creating AI-based models capable of predicting serotonergic activity and selectivity based on ligands' representation by molecular descriptors. Predictive models were developed using Automated Machine Learning provided by Mljar and later analyzed through the SHAP importance analysis, which allowed us to clarify the relationship between descriptors and the effect on activity and what features determine selective affinity for serotonin receptors. Through the experiments, it was possible to highlight the most important features of ligands based on highly efficient models. These features are discussed in this manuscript. The models are available in the additional modules of the SerotoninAI application called "Serotonergic activity" and "Selectivity".

Enhanced Multitask Learning for Hash Code Generation of Palmprint Biometrics.

This paper presents a novel multitask learning framework for palmprint biometrics, which optimizes classification and hashing branches jointly. The classification branch within our framework facilitates the concurrent execution of three distinct tasks: identity recognition and classification of soft biometrics, encompassing gender and chirality. On the other hand, the hashing branch enables the generation of palmprint hash codes, optimizing for minimal storage as templates and efficient matching. The hashing branch derives the complementary information from these tasks by amalgamating knowledge acquired from the classification branch. This approach leads to superior overall performance compared to individual tasks in isolation. To enhance the effectiveness of multitask learning, two additional modules, an attention mechanism module and a customized gate control module, are introduced. These modules are vital in allocating higher weights to crucial channels and facilitating task-specific expert knowledge integration. Furthermore, an automatic weight adjustment module is incorporated to optimize the learning process further. This module fine-tunes the weights assigned to different tasks, improving performance. Integrating the three modules above has shown promising accuracies across various classification tasks and has notably improved authentication accuracy. The extensive experimental results validate the efficacy of our proposed framework.

Interface Development of a Module for the interFoam Solver of the OpenFOAM Package

The results of a study on expanding the list of tasks solved using the graphical shell of the OpenFOAM platform by creating and connecting an additional module for the interFoam solver to the basic version of the graphical shell are presented. The relevance of the problem is formulated, the goal and the list of tasks necessary to achieve it. The features of this solver are studied and described in relation to experiments on numerical modeling of problems in continuum mechanics. The possibility of using open technologies to supplement the source code of the graphical shell with new modules covering a larger range of types of problems in continuum mechanics is analyzed. Graphic materials have been prepared that describe the structure and logic of the module being developed. The technology stack used is presented, and the results of testing the updated version of the graphical shell using the example of one of the classic problems of continuum mechanics are presented. The results of the research were summed up, the tasks completed were noted, the expected practical value of the development and possible ways for further research in this direction were indicated, the repository with the source code of the graphical shell on the GitHub service was updated.

Generation of transmission wave with low AM noise for sub-GHz CW-EPR spectrometer

This study describes a technique to clean amplitude modulation (AM) noise of RF transmission waves, which is used to observe the sub-GHz CW-EPR spectrum. An RF transmitter amplifier that has the function of cleaning AM noise has been developed. Cleaning of the AM noise was owing to saturation of the output at the amplifier. Three stages of the amplifiers in series could effectively suppress the AM noise to about –176 dBc/Hz and –183 dBc/Hz at offset frequency of 10 kHz and 100 kHz, respectively at the carrier frequency of 750 MHz and the output power of 29 dBm. Since phase modulation (PM) noise is suppressed by phase sensitive detection, the AM noise in the transmission is dominant cause of the noise in the sub-GHz CW-EPR absorption spectrum using a reflection bridge, which depends on the quality factor of the resonator and the power of the RF transmission. The additive phase modulation (PM) noise of this amplifier was –171 dBc/Hz at an offset frequency of 100 kHz, which indicated that the frequency modulation (FM) of the transmission wave was not distorted with this amplifier. Therefore, conventional CW-EPR spectrometers that typically require FM for automatic frequency control or automatic tunning control can use this technique to increase sensitivity.

Early transcriptomic responses of rice leaves to herbivory by Spodoptera frugiperda

During herbivory, chewing insects deposit complex oral secretions (OS) onto the plant wound. Understanding how plants respond to the different cues of herbivory remains an active area of research. In this study, we used an herbivory-mimick experiment to investigate the early transcriptional response of rice plants leaves to wounding, OS, and OS microbiota from Spodoptera frugiperda larvae. Wounding induced a massive early response associated to hormones such as jasmonates. This response switched drastically upon OS treatment indicating the activation of OS specific pathways. When comparing native and dysbiotic OS treatments, we observed few gene regulation. This suggests that in addition to wounding the early response in rice is mainly driven by the insect compounds of the OS rather than microbial. However, microbiota affected genes encoding key phytohormone synthesis enzymes, suggesting an additional modulation of plant response by OS microbiota.

Ultralow Electric Current-Assisted Magnetization Switching due to Thermally Engineered Magnetic Anisotropy.

Control of magnetic anisotropy in thin films with perpendicular magnetic anisotropy is of paramount importance for the development of spintronics with ultralow-energy consumption and high density. Traditional magnetoelectric heterostructures utilized the synergistic effect of piezoelectricity and magnetostriction to realize the electric field control of magnetic anisotropy, resulting in additional fabrication and modulation processes and a complicated device architecture. Here, we have systematically investigated the electric current tuning of the magnetic properties of the metallic NiCo2O4 film with intrinsic perpendicular magnetic anisotropy. Ferrimagnetic-to-paramagnetic phase transition has been induced through Joule heating, resulting in a rapid decrease of both magnetic coercivity and moment. An ultralow current density of 2.5 × 104 A/cm2, which is 2 to 3 orders magnitude lower than that of conventional spin transfer torque devices, has been verified to be effective for the control of the magnetic anisotropy of NiCo2O4. Successful triggering of magnetic switching has been realized through the application of a current pulse. These findings provide new perspectives toward the electric control of magnetic anisotropy and design of spintronics with an ultralow driving current density.

Training simulator efficacy in developing thoracic and general surgical skills in a residency programme: a pilot study.

Virtual training simulators have been introduced in several surgical disciplines to improve residents' abilities. Through the use of the LapSim® virtual training simulator (Surgical Science, Göteborg, Sweden), this study aims to plan an effective learning path in minimally invasive thoracic and general surgery. All thoracic and general surgery trainees in their 1st and 2nd year of residency at the University of Insubria were enrolled and randomized into 2 groups: residents undergoing an intensive twice-a-week virtual training programme (group A: n = 8) and those undergoing a once-weekly non-intensive virtual training programme (group B: n = 9). The virtual training programme was divided into 4 modules, each of 12 weeks. In the 1st module, trainees repeated grasping, cutting, clip application, lifting and grasping, and fine dissection exercises during each training session. Seal-and-cut exercise was performed as the initial and final test. Data on surgical manoeuvres (time and on mistakes) were collected; intra- and inter-group comparisons were planned. No significant differences were observed between groups A and B at the 1st session, confirming that the 2 groups had similar skills at the beginning. After 12 weeks, both groups showed improvements, but comparing data between initial and final test, only Group A registered a significant reduction in total time (P-value = 0.0015), left (P-value = 0.0017) and right (P-value = 0.0186) instrument path lengths, and in left (P-value = 0.0010) and right (P-value = 0.0073) instrument angular path lengths, demonstrating that group A acquired greater precision in surgical manoeuvres. Virtual simulator training programme performed at least twice a week was effective for implementing basic surgical skills required for the trainee's professional growth. Additional virtual training modules focused on more complex exercises are planned to confirm these preliminary results.

Generation of a Realistic Synthetic Laryngeal Cancer Cohort for AI Applications

Background: Obtaining large amounts of real patient data involves great efforts and expenses, and processing this data is fraught with data protection concerns. Consequently, data sharing might not always be possible, particularly when large, open science datasets are needed, as for AI development. For such purposes, the generation of realistic synthetic data may be the solution. Our project aimed to generate realistic cancer data with the use case of laryngeal cancer. Methods: We used the open-source software Synthea and programmed an additional module for development, treatment and follow-up for laryngeal cancer by using external, real-world (RW) evidence from guidelines and cancer registries from Germany. To generate an incidence-based cohort view, we randomly drew laryngeal cancer cases from the simulated population and deceased persons, stratified by the real-world age and sex distributions at diagnosis. Results: A module with age- and stage-specific treatment and prognosis for laryngeal cancer was successfully implemented. The synthesized population reflects RW prevalence well, extracting a cohort of 50,000 laryngeal cancer patients. Descriptive data on stage-specific and 5-year overall survival were in accordance with published data. Conclusions: We developed a large cohort of realistic synthetic laryngeal cancer cases with Synthea. Such data can be shared and published open source without data protection issues.

ELONGATED HYPOCOTYL 5a modulates FLOWERING LOCUS T2 and gibberellin levels to control dormancy and bud break in poplar.

Photoperiod is a crucial environmental cue for phenological responses, including growth cessation and winter dormancy in perennial woody plants. Two regulatory modules within the photoperiod pathway explain bud dormancy induction in poplar (Populus spp.): the circadian oscillator LATE ELONGATED HYPOCOTYL 2 (LHY2) and GIGANTEA-like genes (GIs) both regulate the key target for winter dormancy induction FLOWERING LOCUS T2 (FT2). However, modification of LHY2 and GIs cannot completely prevent growth cessation and bud set under short-day (SD) conditions, indicating that additional regulatory modules are likely involved. We identified PtoHY5a, an orthologs of the photomorphogenesis regulatory factor ELONGATED HYPOCOTYL 5 (HY5) in poplar (Populus tomentosa), that directly activates PtoFT2 expression and represses the circadian oscillation of LHY2, indirectly activating PtoFT2 expression. Thus, PtoHY5a suppresses SD-induced growth cessation and bud set. Accordingly, PtoHY5a knockout facilitates dormancy induction. PtoHY5a also inhibits bud-break in poplar by controlling gibberellic acid (GA) levels in apical buds. Additionally, PtoHY5a regulates the photoperiodic control of seasonal growth downstream of phytochrome PHYB2. Thus, PtoHY5a modulates seasonal growth in poplar by regulating the PtoPHYB2-PtoHY5a-PtoFT2 module to determine the onset of winter dormancy, and by fine-tuning GA levels to control bud-break.

Modular organization of shoot system and ontogenesis of Hedysarum branthii Trautv. & C.A. Mey.

The shoot formation of Hedysarum branthii and ontogeny in natural populations on the territory of the Russian Far East were studied from the position of a modular organization. The structure of three categories of modules (elementary, universal and basic) is described. In different ecological-cenotic conditions were revealed two types of shoot. In the structure of the shoot were identifi ed the 15 types of elementary modules, which form 3 types of universal and 3 types of basic modules. The variability of the basic module is leads to the formation of three type of the life form. The life form, which consists of taproot system and elongated, ascending type of shoot is dominant of the H. branthii and in the Gamotion section of the genus Hedysarum. It has been established that a change in the life form occurs with a change in environmental conditions, namely, the altitude above sea level and the type and humidity of the substrate due to the inclusion of two types of additional adaptive elementary modules.

Quasiperiodic Emissions: Fine Structure Corresponding to a Bouncing Wave

AbstractQuasiperiodic (QP) emissions are whistler‐mode electromagnetic waves observed in the Earth's inner magnetosphere whose intensity has a nearly periodic time modulation with typical modulation periods on the order of minutes. Some events exhibit, on top of the main modulation period, an additional fine inner modulation with modulation periods on the order of seconds. We use high‐resolution multi‐component electromagnetic wave data obtained by the Van Allen Probes spacecraft to investigate one such event. Detailed wave analysis demonstrates that the fine inner modulation is due to a wave packet bouncing back and forth between the hemispheres. The presence of a density duct is important for the formation of the event, as demonstrated by the increased ratio of wave power propagating away from the equator (a tentative source region) within the duct. The main QP modulation period corresponds to the plasma number density modulation observed just outside the plasmasphere.

Novel Electric Propulsion System Analysis Method for Electric Vertical Takeoff and Landing Aircraft Conceptual Design

This paper proposes novel enhancements to the electric propulsion system analysis method for electric vertical takeoff and landing (eVTOL) aircraft conceptual design by considering the electrical characteristics of each electrical device in a more accurate manner. To this end, three modules for motor, inverter, and battery analysis are constructed using equivalent circuits and semi-empirical models. First, the motor analysis module is developed using equivalent circuit analysis with the operation control strategy for a permanent-magnet synchronous motor. Second, the inverter analysis module is built using average loss models for the switching and conduction losses. Third, the battery analysis module is improved using the near-linear discharge model to consider the voltage drop during operation. Moreover, additional modules, such as those for calculating the battery stack in series and parallel, as well as regression models for the motor and inverter performance parameters and consideration of the drive system type (direct or indirect, including a reduction gear), are implemented. A comparative study is presented from an analysis and design perspective while changing the drive system type (direct or indirect) and gear ratio, which highlights the importance of modeling electrical characteristics during eVTOL aircraft conceptual design.

Near-infrared H2S telemetry system under non-cooperative target with a novel mixed heterodyne optical structure scheme for high efficiency

A gas leakage telemetry system under non-cooperative target was designed and developed based on laser absorption spectroscopy technology, targeting the combination frequency absorption band of Hydrogen sulfide (H2S) gas molecules. A distributed feedback (DFB) laser centered at 1578 nm was adopt, with mature packaging process and high market maturity. The returned signal of diffuse reflection is extremely weak, decreasing exponentially with the increase of telemetry distance, and direct detection technology is difficult to achieve ideal performance at telemetry distance and lower detection limit. Thus, heterodyne detection technology was introduced to significantly amplify diffuse reflection signal and improve the signal-to-noise ratio (SNR) of telemetry system. For further detection sensitivity, wavelength modulation spectroscopy (WMS) technology was combined to suppress background noise through the electrical modulation of DFB laser. Alternating signal was applied to the laser to alter its output characteristics during the process of laser oscillation. Compared with exterior modulation methods, this interior modulation method does not require additional modulator, which avoids optical loss and is more economical. Combining the advantages of fiber structure and space optical path structure, a novel mixed heterodyne optical structure which was more applicable to this system was independently designed after the specific analysis of main noise sources. Then, the effect of heterodyne detection technology was verified in contrast of direct detection technology. Sensor performance was validated through calibration and stability experiments. The limit of detection (LoD) reached a minimum of 3.44 parts-per-million (ppm·m) with a 98 s averaging time. And the packaged system was fully capable of detecting the hazardous critical concentration of H2S and acting as an alarm mechanism.

Associations between violent crime inside and outside, air temperature, urban heat island magnitude and urban green space

There are more incidents of violence in summer and on hot days, a trend likely to be exacerbated by climate change. Urban areas experience additional temperature modulation due to the urban form, however, to date, no studies have considered the effect of the urban heat island (UHI) or green space with respect to the temperature–violence relationship. This study modelled the relationship between the number of daily violent crime incidents that occurred inside or outside between July 2013 and June 2018, and the average surface UHI or percentage greencover (including grasses, shrubs and trees) within each local government area in Greater Sydney, Australia. Panelised negative binomial time series regression models indicated that the violent crime rate was associated with higher surface UHI for crimes committed outside (p = 0.006) but not inside (p = 0.072). Greater percentage of all vegetation was associated with significantly lower rates of violent crime committed outside (p = 0.011) but was not associated with violent crimes committed inside (p = 0.430). More socio-economic disadvantage was associated with higher rates of violent crime committed inside (p = 0.002) but not outside (p = 0.145). Greater temperature was non-linearly associated with higher rates of violent crime committed both inside and outside (p < 0.001). The findings of this study are important because both violence and heat exposure are critical health issues and will be stressed by urbanisation and climate change. The expansion of green space and/or reduction in UHI may mitigate these effects.

Improving Parent-Child Interaction and Reducing Parental Violent Discipline – a Multi-Informant Multi-Method Pilot Feasibility Study of a School-Based Intervention

Abstract Purpose Globally, many children are exposed to violent discipline in multiple settings. Interventions to prevent violent discipline are therefore highly needed. In the present study, the feasibility of the intervention Interaction Competencies with Children – for Parents (ICC-P), an additional module of a school-based intervention for teachers, was tested. The intervention aims to prevent violent discipline by changing attitudes towards such method and fostering supportive adult-child interaction through non-violent interaction skills. Methods In total, 164 parents (Mage= 39.55, range = 24 70, 72.3% female) from four public secondary schools in Tanzania participated in a four-day training conducted by six trainers (Mage= 44.67, range = 40–47, 50% female). Using a One-Group Pre-Post design, we measured the feasibility and preliminary effectiveness of the intervention qualitatively and quantitatively. Parents were assessed via self-administered questionnaires before and six weeks after the intervention. Trainers rated the implementation of every workshop session. Results Based on descriptive statistics and Classical Content Analysis, implementing trainers and participants rated ICC-P as feasible. Participants indicated a high need for such interventions and showed high acceptance. They were able to integrate core aspects of the intervention in their daily interactions with children. Using t-tests, ICC-P proved to be preliminarily effective; parents reported applying less violent discipline and holding more critical attitudes about such measures after the intervention. Conclusion ICC-P is feasible intervention that showed initial signs of effectiveness. We recommend combining the parents’ training module with the teachers’ module to prevent violence in multiple settings.