Flexible Combination Research Articles

Analysis of CNN-based remote-PPG to understand limitations and sensitivities.

Deep learning based on convolutional neural network (CNN) has shown promising results in various vision-based applications, recently also in camera-based vital signs monitoring. The CNN-based photoplethysmography (PPG) extraction has, so far, been focused on performance rather than understanding. In this paper, we try to answer four questions with experiments aiming at improving our understanding of this methodology as it gains popularity. We conclude that the network exploits the blood absorption variation to extract the physiological signals, and that the choice and parameters (phase, spectral content, etc.) of the reference-signal may be more critical than anticipated. The availability of multiple convolutional kernels is necessary for CNN to arrive at a flexible channel combination through the spatial operation, but may not provide the same motion-robustness as a multi-site measurement using knowledge-based PPG extraction. We also find that the PPG-related prior knowledge may still be helpful for the CNN-based PPG extraction, and recommend further investigation of hybrid CNN-based methods that include prior knowledge in their design.

A dynamic weighting adjustment algorithm for hybrid gray model based on artificial neural network

ABSTRACTHybrid gray model is a combination of gray model and other mathematical models to obtain a high precision prediction. It has demonstrated good performance in many applications. However, there is little discussion on the optimal combination of weights among these models. For this purpose, this paper proposes a dynamic weighting hybrid gray model to provide a flexible combination to adapt to both stable and unstable time series. Short, medium and long term modeling numbers are used to verify the reliability of the proposed method. Two illustrative examples are shown to compare the prediction accuracy of the proposed method with that of the classical hybrid gray model and ANN model. Results show that the proposed model is better and is more adaptable to a time series with rapid changes.

Analytical classical density functionals from an equation learning network.

We explore the feasibility of using machine learning methods to obtain an analytic form of the classical free energy functional for two model fluids, hard rods and Lennard-Jones, in one dimension. The equation learning network proposed by Martius and Lampert [e-print arXiv:1610.02995 (2016)] is suitably modified to construct free energy densities which are functions of a set of weighted densities and which are built from a small number of basis functions with flexible combination rules. This setup considerably enlarges the functional space used in the machine learning optimization as compared to the previous work [S.-C. Lin and M. Oettel, SciPost Phys. 6, 025 (2019)] where the functional is limited to a simple polynomial form. As a result, we find a good approximation for the exact hard rod functional and its direct correlation function. For the Lennard-Jones fluid, we let the network learn (i) the full excess free energy functional and (ii) the excess free energy functional related to interparticle attractions. Both functionals show a good agreement with simulated density profiles for thermodynamic parameters inside and outside the training region.

"Off/on" fluorescence imaging-guided cancer diagnosis and multi-modal therapy.

An efficient theranostic nanoplatform responding to tumour microenvironments with characters of simple and flexible combinations owns great potential in cancer diagnosis and therapy. Herein, a series of triblock copolymers, mPEG-b-PDPA-b-P(nBMA-r-cystamine) (EPB), were synthesized and among them, the structure of EPB-3 was optimized for both fluorescence imaging-guided cancer diagnosis and multi-modal therapy with good biocompatibility. (1) The self-assembled nanoparticles of EPB-3-ICG1 obtained by conjugating one ICG on EPB-3 via S-S bonds effectively performed reduction-sensitive OFF/ON fluorescence signal transition, thus inducing tumour cell-specific amplified fluorescence imaging in vitro and in vivo. (2) By entrapping Au nanorods into the co-assembled NPs of EPB-3 and EPB-3-ICG1, EPB-3-ICG1@Au NPs could synchronously induce strong tumour fluorescence imaging and high local photothermal effect, indicating the potential of imagine-guided photothermal therapy. (3) EPB-3 NPs could efficiently co-load paclitaxel (PTX) and ICG to form stable EPB-3@PTX@ICG NPs, which provided long periods of intracellular pH-sensitive sustainable drug release and highly enhanced apoptosis of 4T1 cells in vitro by the chemo-photothermal effect. Excitingly, a single intravenous injection of EPB-3@PTX@ICG NPs followed by a one-time local near-infrared light (NIR, 808 nm) irradiation treatment for 10 min could lead to significant inhibition of tumour growth, avoiding tumor metastasis and extending the survival of mice. All the above-mentioned results suggest that EPB-3 provides a nanoplatform with the characters of simple structure, convenience of use and flexible combination, holding potential for multi-modal diagnosis and therapy.

Capacity Enhancement of a Radial Distribution Grid Using Smart Transformer

Distributed generation (DG) and electric vehicle (EV) are increasingly installed in the modern microgrids (MGs). This demands for an improvement in the power handling capacity of the existing distribution feeders. The power handling capacity of a feeder is limited by the voltage magnitude violation caused by overloading. The smart transformer (ST) is proposed as an effective solution for improving the performance of modern MG. This paper utilizes the power management capabilities of the ST with an integrated battery energy storage system (BESS) for improving the power handling capacity of a medium voltage (MV) 5-bus radial feeder. The proposed method demonstrates the ST's capability to control the active and/or reactive power flow at the MV grid for maintaining the steady-state voltage magnitude within the grid-specified limit. A sensitivity analysis based design study is proposed to determine the power rating requirements for different ST and BESS converters during distinct power control scenarios. The analysis shows the capability of ST in injecting/absorbing active and reactive powers in flexible combinations for achieving the desired voltage magnitudes. The findings are verified through simulation and experimental results.

Psychoeducation as an element of psychosocial interventions in a complex system of medic-psychological support for a family, where a patient with endogenous mental disorder lives

The biopsychosocial model of therapy for endogenous mental disorders involves a flexible combination of psychopharmacotherapy with psychosocial interventions. Psychoeducation is one of the most important components of psychosocial interventions in a multifaceted system of psychosocial rehabilitation. The primary task of psychoeducation is to provide patients and their family caregivers with realistic knowledge about mental disorder, on the basis of which the patient and his or her family members can get more control over the symptoms of the disease. The aim of our work was from the standpoint of a systematic approach, based on the study of clinical and psychological manifestations and risk factors for the development of pathological functioning in a family, where a patient with endogenous mental disorder lives, to develop, substantiate scientifically and introduce a psychoeducational module as an element of psychosocial interventions in complex system of medical -psychological support. To achieve this goal, according to the principles of bioethics and medical deontology, a comprehensive examination of 243 patients with endogenous mental disorders (168 patients with paranoid schizophrenia, 75 patients with affective disorders) and 243 family caregivers was performed. The work was done in three stages: during the the first stage we examined patients and their family caregivers. During the second stage, an in-depth study of psycho-emotional, individual-psychological, interpersonal-communicative and psychosocial predictors of reducing the adaptive capacity of the family was performed. The third stage included scientific substantiation, development and implementation of an appropriate comprehensive system of medical and psychological support for the families, based on the analysis of data obtained during the previous stages. The proposed system of medical and psychological support has shown its effectiveness in improving the adaptive capacity of the family, where patients with endogenous mental disorders live, and can be recommended for further implementation in appropriate medical institutions.

Combined Design and Process Planning for Incremental Manufacturing

Incremental Manufacturing (IM) is a hybrid-manufacturing concept where standardized base parts are produced in large volume production first, then assembled and finalized by additive and subtractive manufacturing. The concept allows for a flexible combination of different manufacturing processes. These new possibilities result in a high degree of freedom in both parts design and process planning, which must be used purposefully and restricted in a manageable way with regard to production-related restrictions. This paper presents an approach for evaluating different production routes of an IM designed part.

Інноваційно-активний університет: підґрунтя виникнення та особливості функціонування

The article aims at defining the essence of the "innovatively active university" concept as well as the key characteristics and structural components of such a university; substantiating the management framework of educational innovations taking place at a university and providing a flexible combination of teaching, research and entrepreneurship. The article considers the regulatory support present in Ukraine as for the functioning of the education system, including all its levels and subsystems, the activities of various educational institutions, public or private, irrespective of their attendance modes. The analysis of definitions of the "innovative university" term suggested by various scientists has been carried out thus allowing the author to consider this concept within the functional approach. It has been determined that the given terms are similar in their meaning, but still the difference between them is rather significant. For a deeper understanding of the concept, decomposition of the abovementioned definitions into the main components has been made according to the following points: the essential component of the concept; activity areas of the innovative university; the mission of an innovative university. The "innovatively active university" has been defined as an entrepreneurial organization that possesses the resource readiness to promote accelerated social and economic development through intensive transfer of knowledge and technology generated at the university due to partnership relations established with stakeholders (labor market actors, government and NGOs). Models of the innovative university and their distinctive features have been suggested. The main characteristics of the innovative university that contribute to the development of new systems and management processes have been outlined. The ways of managing innovative universities have been considered and the management framework of educational innovations at a university has been offered. Prospects for further research in this area lie in the following: analyzing the competitive strength of universities around the world to determine the internal and external features of an innovatively active university, helping it to achieve and maintain a high level of competitiveness and attractiveness.

ADVISOR: An Adjustable Framework for Test Oracle Automation of Visual Output Systems

Test oracles differentiate between the correct and incorrect system behavior. Automation of test oracles for visual output systems mainly involves image comparison, where a snapshot of the output is compared with respect to a reference image. Hereby, the captured snapshot can be subject to variations such as scaling and shifting. These variations lead to incorrect evaluations. Existing approaches employ computer vision techniques to address a specific set of variations. In this article, we introduce ADVISOR, an adjustable framework for test oracle automation of visual output systems. It allows the use of a flexible combination and configuration of computer vision techniques. We evaluated a set of valid configurations with a benchmark dataset collected during the tests of commercial digital TV systems. Some of these configurations achieved up to 3% better overall accuracy with respect to state-of-the-art tools. Further, we observed that there is no configuration that reaches the best accuracy for all types of image variations. We also empirically investigated the impact of significant parameters. One of them is a threshold regarding image matching score that determines the final verdict. This parameter is automatically tuned by offline training. We evaluated runtime performance as well. Results showed that differences among the ADVISOR configurations and state-of-the-art tools are in the order of seconds per image comparison.

Real-Time Multi-person Motion Capture from Multi-view Video and IMUs

A real-time motion capture system is presented which uses input from multiple standard video cameras and inertial measurement units (IMUs). The system is able to track multiple people simultaneously and requires no optical markers, specialized infra-red cameras or foreground/background segmentation, making it applicable to general indoor and outdoor scenarios with dynamic backgrounds and lighting. To overcome limitations of prior video or IMU-only approaches, we propose to use flexible combinations of multiple-view, calibrated video and IMU input along with a pose prior in an online optimization-based framework, which allows the full 6-DoF motion to be recovered including axial rotation of limbs and drift-free global position. A method for sorting and assigning raw input 2D keypoint detections into corresponding subjects is presented which facilitates multi-person tracking and rejection of any bystanders in the scene. The approach is evaluated on data from several indoor and outdoor capture environments with one or more subjects and the trade-off between input sparsity and tracking performance is discussed. State-of-the-art pose estimation performance is obtained on the Total Capture (mutli-view video and IMU) and Human 3.6M (multi-view video) datasets. Finally, a live demonstrator for the approach is presented showing real-time capture, solving and character animation using a light-weight, commodity hardware setup.

Application of User Experience and Design Thinking to the Construction of a Class Assistance System for Hearing- and Speech-Impaired People

According to a study by the Department of Statistics, Ministry of Health and Welfare, Taiwan, a total of 15,145 people had a voice function impairment or speech function impairment, and a total of 122,533 people had a hearing impairment by the second quarter (Q2) of 2018. Hearing and speech are the most commonly used sensory functions for communication. Therefore, a person will suffer unimaginable difficulties with the loss of both functions during her/his life. The problems that a handicapped student might encounter during school are often neglected. The way a teacher teaches during a class and how a student interacts with others should be highlighted. The deficiency in assistive devices for hearing- and speech-impaired people is due to several factors. In fact, there is insufficient information and not enough researchers. The translation of language by foreign assistive devices is more rigorous; however, domestic resources for assistive devices for hearing- and speech-impaired people are limited. No relevant information on learning through assistive devices is available. The subject examined in this study was a hearing-impaired student in her junior year, who lost speech capability after an ear stroke. This study examined the problems that she encountered during her learning in a class with average students. After a literature review and the investigation of the case study, user experience and design thinking approaches were implemented for the development of an assistive device. A class learning assistive system, which is based on a simple innovative design and a highly flexible combination of elements, was created. After a series of experimental verifications and improvements, the results indicated that this learning assistive system could effectively enhance the research subject’s confidence and autonomy during learning in class.

Традиции и информация

Традиции и информация

Improvement of education for clinical nutrition interns based on Joint Commission International standards

Huashan Hospital became the first Academic Medical Center Hospital accredited by Joint Commission International (JCI) in 2013. The Department of Clinical Nutrition has constantly improved internship education through the introduction of JCI standards. Based on the flexible combination of teaching and practicing modules, clinical nutrition knowledge is closely integrated with professional dietitian skills, and in-depth practical training helps students to acquire much more experience of the occupation as a dietitian. Teachers think highly of students' initiative and knowledge conversion ability, and in the recent five years, undergraduate projects have been enhanced in both quantity and quality. Key words: Joint Commission International standards; Clinical nutrition; Intern education

Ultra-broadband metamaterial absorber in long wavelength Infrared band based on resonant cavity modes

A wavelength-selective, ultrathin, broadband infrared metal–insulator–metal (MIM) absorber with multi-sized Ti–Ge cubes is proposed. By flexibly combining different sizes of Ti–Ge cubes, stacking upward and reasonably filling the border, all the resulting 3 structures that we propose show an average absorptivity of at least 90% over a wide spectral range in the infra-red, extending from 6. 3μm to 14.8μm. The absorptivity is further increased up to 99% when considering narrower spectral ranges such as 7.96-8.34μm and 11.02-11.75μm wavebands, which is superior than the previous work in long wavelength infrared band based on the MIM structure. Those characteristics vary depending on the 3 designs considered and studied in this work: single-layer structure, double-layer structure and modified structure. The intrinsic strong energy dissipation caused by highly lossy metal Ti (here Titanium) and excited low-Q cavity modes are key factors contributing to such efficient broadband absorption. The polarization and angle insensitivity are demonstrated by analysing the absorption performance with oblique incidences for both transverse electric wave (TE) and transverse magnetic wave (TM). Moreover, flexible combinations of different resonators allow trade-offs between the absorption bandwidth and the absorbance, which makes the operating waveband of the metamaterial absorber adjustable through proper design. The proposed broadband absorbers have many potential applications, including microbolometers, thermal emitters, and plasmonic sensors.

Combination of a flexible electrode optimizer with a new forward modeling technique in 3D survey

In this research a flexible electrode array's combination device is used for 2D/3D resistivity data acquisition. This system can be used for applying any possible array over a survey electrode netw...

Containment control of high‐order multi‐agent systems with heterogeneous uncertainties, dynamic leaders, and time delay

AbstractThis paper deals with containment control of multi‐agent systems where high‐order canonical controllable linear systems with heterogeneous uncertainties are considered as the agents' dynamics. The leaders are dynamic with polynomial‐type trajectories. In this paper, the control protocol is smooth and it is constructed from a flexible combination of local state feedback controller and a PI‐type protocol. Utilizing this protocol, containment without suffering from chattering phenomena and consideration of any bound on the agents' states is achieved. In addition, the flexibility of this protocol helps the designer to reduce or even discard the inter‐agent communication. Recalling the significant effects of time delay on the performance and stability of multi‐agent systems, in this paper input time delay is also considered. Following a Lyapunov‐Krasovskii stability approach, two linear matrix inequalities (LMIs) are derived, and then solving the LMIs' gains of the protocol are attained such that asymptotic stability of the multi‐agent system under switching graph topology is guaranteed. Finally, the feasibility of the LMIs and the validity of the theoretical results are demonstrated through some numerical examples.

An efficient artificial neural network for damage detection in bridges and beam-like structures by improving training parameters using cuckoo search algorithm

This paper presents a new approach for damage detection in structures by applying a flexible combination based on an artificial neural network (ANN) and cuckoo search (CS) algorithm. ANN has become one of the most powerful tools employing computational intelligence techniques to tackle complex problems in numerous fields. However, due to the application of backpropagation algorithms based on gradient descent, a major drawback of ANN is the common problem of local minima that acts as a great hindrance to the search for the best solution. To overcome this disadvantage, we propose to combine ANN with evolutionary algorithms based on global search techniques. This paper employs CS to improve ANN training parameters (weight and bias) by minimizing the difference between real and desired outputs and then using these parameters to generate the network. Two numerical models, comprising a steel beam calibrated using experimental measurements and a large-scale truss bridge, are used to assess the robustness of the proposed approach. The results demonstrate that ANN combined with CS (ANN-CS) is accurate and requires a lower computational time than ANN, and evolutionary algorithm (EA) alone in terms of structural damage localization and quantification.

Flexible combination of reward information across primates.

A fundamental but rarely contested assumption in economics and neuroeconomics is that decision-makers compute subjective values of risky options by multiplying functions of reward probability and magnitude. In contrast, an additive strategy for valuation allows flexible combination of reward information required in uncertain or changing environments. We hypothesized that the level of uncertainty in the reward environment should determine the strategy used for valuation and choice. To test this hypothesis, we examined choice between risky options in humans and monkeys across three tasks with different levels of uncertainty. We found that whereas humans and monkeys adopted a multiplicative strategy under risk when probabilities are known, both species spontaneously adopted an additive strategy under uncertainty when probabilities must be learned. Additionally, the level of volatility influenced relative weighting of certain and uncertain reward information and this was reflected in the encoding of reward magnitude by neurons in the dorsolateral prefrontal cortex.

Focal Plane Array Based on the InGaAs/InP Heterostructure for 3D Imaging in Short-Wave IR Range

A Russian-made focal plane array (FPA) for gated imager working in the short-wave IR range is presented. The FPA contains an array of p–i–n photodiodes based on a 320 × 256 InGaAs/InP heterostructure with a pitch of 30 µm, readout integrated circuit, thermoelectric cooler, and a sealed housing with a sapphire window. A distinctive feature of the FPA is related to functioning in four modes: passive, gated 3D, gated 2D, and asynchronous binary. Flexible combinations of such modes can be used to obtain maximum data on an object under observation. Range data that are formed in each pixel of the FPA and the signal intensities make it possible to synthesize 3D images of objects. The main parameters of the FPA and IR images obtained in different modes are demonstrated. Calculated ranging results are presented versus power and divergence of laser radiation. The main sources of errors in calculations of range resolution are considered.

Fixed versus flexible combination of GLP-1 receptor agonists with basal insulin in type 2 diabetes: A retrospective multicentre comparative effectiveness study.

Background and AimsThe combination of basal insulin (BI) and GLP‐1 receptor agonists (GLP‐1RAs) is a rational and effective therapy for patients with uncontrolled type 2 diabetes (T2D). We compared the effectiveness of fixed and flexible BI/GLP‐1RA combinations using routinely accumulated clinical data.MethodsThis was a retrospective, multicentre, real‐world study concerning T2D patients initiating a fixed or flexible BI/GLP‐1RA combination (NCT03959865). The primary endpoint was change in HbA1c. Secondary endpoints were changes in body weight, fasting plasma glucose (FPG) and systolic blood pressure (SBP). Confounding was addressed by propensity score matching (PSM) or multivariable adjustment (MVA).ResultsA total of 609 patients were included in the study, 131 in the fixed group and 478 in the flexible group. The two groups differed in terms of diabetes duration, body weight and concomitant medications. After 5.7 months, observed HbA1c reductions were 0.6% and 0.8%, and body weight reductions were 2.8 kg and 1.2 kg in the flexible and fixed groups, respectively. Following PSM, HbA1c declined similarly in the two groups, whereas reduction in body weight was significantly in favour of the flexible combination. Findings were robust in sensitivity analyses, with the exception that, with MVA, a significantly higher reduction in HbA1c was detected in the fixed group. Final doses of BI were higher in the fixed group, whereas those of GLP‐1RA were higher in the flexible group.ConclusionsIn routine specialist care, initiation of the fixed or flexible BI/GLP‐1RA combination allowed similar improvement in glycaemic control, but greater weight loss was observed with the flexible combination. This difference reflected dosages of BI and GLP‐1RAs.