- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
A computer-aided diagnosis system for HEp-2 fluorescence intensity classification
摘要： Background and objective: The indirect immunofluorescence (IIF) on HEp-2 cells is the recommended technique for the detection of antinuclear antibodies. However, it is burdened by some limitations, as it is time consuming and subjective, and it requires trained personnel. In other fields the adoption of deep neural networks has provided an effective high-level abstraction of the raw data, resulting in the ability to automatically generate optimized high-level features. Methods: To alleviate IIF limitations, this paper presents a computer-aided diagnosis (CAD) system classifying HEp-2 fluorescence intensity: it represents each image using an Invariant Scattering Convolutional Network (Scatnet), which is locally translation invariant and stable to deformations, a characteristic useful in case of HEp-2 samples. To cope with the inter-observer discrepancies found in the dataset, we also introduce a method for gold standard computation that assigns a label and a reliability score to each HEp-2 sample on the basis of annotations provided by expert physicians. Features by Scatnet and gold standard information are then used to train a Support Vector Machine. Results: The proposed CAD is tested on a new dataset of 1771 images annotated by three independent medical centers. The performances achieved by our CAD in recognizing positive, weak positive and negative samples are also compared against those obtained by other two approaches presented so far in the literature. The same system trained on this new dataset is then tested on two public datasets, namely MIVIA and I3Asel. Conclusions: The results confirm the effectiveness of our proposal, also revealing that it achieves the same performance as medical experts.
关键词： HEp-2 samples,Deep learning,Invariant Scattering Convolutional Networks,Computer-aided diagnosis,Indirect immunofluorescence
更新于2025-09-23 15:21:21
-
Non-invasive Seizure Localization with Ictal Single-Photon Emission Computed Tomography is Impacted by Preictal/Early Ictal Network Dynamics
摘要： More than one third of children with epilepsy have medically intractable seizures. Promising therapies, including targeted neurostimulation and surgery, depend on accurate localization of the epileptogenic zone. Ictal perfusion Single-Photon Emission Computed Tomography (SPECT) can localize the seizure focus noninvasively, with comparable accuracy to that of invasive EEG. However, multiple factors including seizure dynamics may affect its spatial specificity. Methods: Using subtracted ictal from interictal SPECT and scalp EEG from 118 pediatric epilepsy patients (40 of whom had surgery after the SPECT studies), information theoretic measures of association and advanced statistical models, this study investigated the impact of preictal and ictal brain network dynamics on SPECT focality. Results: Network dynamics significantly impacted the SPECT localization ～30 s before to ～45 s following ictal onset. Distributed early ictal connectivity changes, indicative of a rapidly evolving seizure, were negatively associated with SPECT focality. Spatially localized connectivity changes later in the seizure, indicating slower seizure propagation, were positively associated with SPECT focality. In the first ～60 s of the seizure, significantly higher network connectivity was estimated in an area overlapping with the area of hyperperfusion. Finally, ～75% of patients with Engel class 1a/1b outcomes had SPECTs that were concordant with the resected area. Conclusion: Slowly evolving seizures are more likely to be accurately imaged with SPECT, and the identified focus may overlap with brain regions where significant topological changes occur. Significance: Measures of preictal/early ictal network dynamics may help optimize the SPECT localization, leading to improved surgical and neurostimulation outcomes in refractory epilepsy.
关键词： ictal SPECT,noninvasive source localization,brain networks,Epilepsy
更新于2025-09-23 15:21:21
-
Alien Wavelengths Over Optical Transport Networks
摘要： To reduce network infrastructure cost, network operators want to integrate interoperable and open transponders, since these transponders allow application of the latest technologies over legacy networks at a competitive price. This process of using third-party transponders in a given network is commonly called “alien wavelength” support. Yet, moving toward interoperability raises several challenges: incompatible physical parameters, vendor lock-in, and proprietary software. Manual set up of alien wavelengths has been demonstrated, but automating this process is essential to enable alien wavelength operation in the field. This paper sheds light on the prevailing literature on the concept of alien wavelengths, taking into account many challenges that accompany the concept on its various levels. We particularly focus on several approaches proposed in the literature: protocol message translation, controller cooperation, and open line systems, including our own proposals, an OpenROADM-based approach and a RSVP-TE-based approach. We analyze these approaches with respect to several criteria: applicability to legacy equipment, added operational cost, and the offered level of interoperability and openness. According to these characteristics, we note that, even though some non-interoperable approaches might be applicable to legacy equipment and easy to maintain, they remain infeasible with advanced alien scenarios (high level of interoperability). Finally, a combination has to be made between at least two of the given approaches in order to get the optimal solution.
关键词： Optical networks,Network interoperability,Alien wavelength
更新于2025-09-23 15:21:21
-
[Advances in Intelligent Systems and Computing] Recent Findings in Intelligent Computing Techniques Volume 709 (Proceedings of the 5th ICACNI 2017, Volume 3) || Optimal Approach for Image Recognition Using Deep Convolutional Architecture
摘要： In the recent time, deep learning has achieved huge popularity due to its performance in various machine learning algorithms. Deep learning as hierarchical or structured learning attempts to model high-level abstractions in data by using a group of processing layers. The foundation of deep learning architectures is inspired by the understanding of information processing and neural responses in human brain. The architectures are created by stacking multiple linear or nonlinear operations. The article mainly focuses on the state-of-the-art deep learning models and various real-world application-speci?c training methods. Selecting optimal architecture for speci?c problem is a challenging task; at a closing stage of the article, we proposed optimal approach to deep convolutional architecture for the application of image recognition.
关键词： Deep neural networks,Image recognition,Image processing,Transfer learning,Convolutional neural networks,Deep learning
更新于2025-09-23 15:21:01
-
Evaluation of electrical efficiency of photovoltaic thermal solar collector
摘要： In this study, machine learning methods of artificial neural networks (ANNs), least squares support vector machines (LSSVM), and neuro-fuzzy are used for advancing prediction models for thermal performance of a photovoltaic-thermal solar collector (PV/T). In the proposed models, the inlet temperature, flow rate, heat, solar radiation, and the sun heat have been considered as the input variables. Data set has been extracted through experimental measurements from a novel solar collector system. Different analyses are performed to examine the credibility of the introduced models and evaluate their performances. The proposed LSSVM model outperformed the ANFIS and ANNs models. LSSVM model is reported suitable when the laboratory measurements are costly and time-consuming, or achieving such values requires sophisticated interpretations.
关键词： hybrid machine learning model,Renewable energy,photovoltaic-thermal (PV/T),least square support vector machine (LSSVM),adaptive neuro-fuzzy inference system (ANFIS),neural networks (NNs)
更新于2025-09-23 15:21:01
-
Dual-Polarization Frequency Selective Rasorber With Independently Controlled Dual-Band Transmission Response
摘要： It is of significant importance for any classification and recognition system, which claims near or better than human performance to be immune to small perturbations in the dataset. Researchers found out that neural networks are not very robust to small perturbations and can easily be fooled to persistently misclassify by adding a particular class of noise in the test data. This, so-called adversarial noise severely deteriorates the performance of neural networks, which otherwise perform really well on unperturbed dataset. It has been recently proposed that neural networks can be made robust against adversarial noise by training them using the data corrupted with adversarial noise itself. Following this approach, in this paper, we propose a new mechanism to generate a powerful adversarial noise model based on K-support norm to train neural networks. We tested our approach on two benchmark datasets, namely the MNIST and STL-10, using muti-layer perceptron and convolutional neural networks. Experimental results demonstrate that neural networks trained with the proposed technique show significant improvement in robustness as compared to state-of-the-art techniques.
关键词： robustness,generalization,convolutional neural networks,adversarial,K-Support norm
更新于2025-09-23 15:21:01
-
[IEEE 2019 International Topical Meeting on Microwave Photonics (MWP) - Ottawa, ON, Canada (2019.10.7-2019.10.10)] 2019 International Topical Meeting on Microwave Photonics (MWP) - Heterogeneous multicore fiber for optical beamforming
摘要： We experimentally demonstrate, for the first-time to our knowledge, optical beamforming for microwave phased array antennas implemented with a heterogeneous multicore fiber link. The multicore fiber has been engineered to act as an optical sampled true time delay that allows to implement radiofrequency signal processing in a distributed way. It comprises 7 trench-assisted cores where each core is fabricated with different dimensions and core dopant concentration, as to feature a different group delay and chromatic dispersion behavior. We emulated different radio beamsteering scenarios where the beam-pointing angle is modified by tuning the optical wavelength in a 20-nm range, while squint-beam effects are avoided.
关键词： Microwave Photonics,optical beamforming networks,multicore fibers
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - PV Curtailment Analysis to Improve Utilization of Hosting Capacity in California
摘要： Healthcare data are becoming increasingly important in the life of people. By utilizing healthcare data in a proper and secure manner, the elderly may avoid some sudden diseases, whereas young people can monitor their health condition. In the hospital, for certain sizes of detection objects, an effective method of data transmission becomes very signi?cant. In view of the movement of patients in the hospital, we introduce a type of network called incompletely predictable networks to describe such scenarios. The patients move in a certain trend or are only active in a certain limited range. To achieve high performance when transmitting healthcare data in such networks, a novel protocol called the direction density-based secure routing protocol is proposed in this paper. Both the moving direction and the in?uence of node group movement are considered. The novel protocol innovatively takes the density of the node moving direction into consideration, which makes full use of the relationships among the moving individuals. Moreover, the design of the secure routing with authenticated message transmission ensures secure healthcare data communication. The simulation shows that our protocol achieves a high packet delivery ratio with low overhead and end-to-end delay.
关键词： Direction density,secure routing protocol,healthcare data,incompletely predictable networks
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Reducing CPV Materials Cost Through Multistage Concentration
摘要： This paper considers an energy-limited cognitive relay network, where a secondary transmitter (ST) assists to forward the traf?c from a primary transmitter (PT) to a primary receiver (PR), in exchange for serving its own secondary receiver (SR) in the same frequency. The multiple-antenna ST is assumed to be energy-constrained and powered by both information ?ow from source (PT) and dedicated energy streams from destinations (PR and SR), which is called a destination-aided wireless power transfer (DWPT) scheme. Then, the relay processing matrix, cognitive beamforming vector, and power splitter are jointly designed to maximize the rate of secondary users under the energy causality constraint and the constraint that the demanded rate of primary users is satis?ed. For the perfect channel state information (CSI) case, by adopting the semi-de?nite relax technique and the Charnes–Cooper transformation, the global optimal solution is given. To reduce the complexity, matrix decomposition, zero forcing scheme, and dual method are jointly employed to derive a suboptimal solution. For the imperfect CSI case, the S-procedure is used to transform the worst case robust problem into a tractable semi-de?nite program. Simulation results reveal that our proposed DWPT scheme is greatly preferred for both perfect and imperfect CSI cases when ST is close to PR/SR.
关键词： cognitive relay networks,power splitting,Wireless power transfer,semi-de?nite program,beamforming design
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Inorganic cesium-lead mixed halide perovskite p-i-n solar cells deposited using layer-by-layer vacuum deposition technique
摘要： Heterogeneous cellular networks (HetCNets) offer a promising solution to cope with the current cellular coverage crunch. Due to the large transmit power disparity, while following maximum power received (MPR) association scheme, a larger number of users are associated with macro-cell BS (MBS) than small-cell BSs (SBSs). Therefore, an imbalance load arrangement takes place across the HetCNets. Hence, using cell range expansion-based cell association, we can balance the load across the congested MBS. However, using MPR association scheme, users’ of?oading leads to two challenges: 1) macro-cell interference, in which the MBS interferes with the of?oaded users, and 2) coupled downlink-uplink cell association, in which a random user associates with a single tier’s base station (BS) both in uplink (UL) and downlink (DL) directions. This paper aims to address these problems while considering a two-tier scenario consisting of small-cell and macro-cell tiers. For the MBS interference mitigation, we employ a reverse frequency allocation (RFA) scheme. Besides coupled DL–UL association (Co-DUA), this paper also highlights the notion of decoupled DL–UL association (De-DUA). In De-DUA, a random user associates with two different tiers’ BSs, i.e., with one tier’s BS in the DL direction and with the other tier’s BS in the UL direction. Our results illustrate that, in comparison with the Co-DUA, De-DUA with RFA employment achieves a better coverage performance.
关键词： small-cell BSs,decoupled downlink-uplink association,Heterogeneous cellular networks,coverage performance,reverse frequency allocation
更新于2025-09-23 15:21:01