OPTORER: A Dynamic Routing and Touring Service for Indoors and Outdoor Tours.

This paper introduces a new routing and touring service both for outdoor and indoor places of touristic and cultural interest designed to be used in the wider area of Attica, Greece. This service is the result of the work performed in OPTORER (OPTORER: OPtimal rouTing and explOration of touRistic and cultural arEas of interest within Attica given personalized adaptive preferences, promoted underlying purpose, and interactive experience), project, and it aspires to offer a range of innovative and thematic routes to several specified points of interest in the selected area of Attica, encouraging the combination of indoor and outdoor routes in a single tour. The aim is to optimize the user experience while promoting specific, user-centric features, with safety and social welfare being a priority for every designed tour, resulting in enhancing the touristic experience in the area. Using a common smartphone device, as well as common wearable devices (i.e., smartwatches), the OPTORER service will provide an end-to-end solution by developing the algorithms and end-user applications, together with an orchestration platform responsible for managing, operating, and executing the service that produces and presents to the end user results derived from solving dynamically complex optimization problems.

Novel Modification of Integrated Optimization Method for Sensor's Communication in Wi-Fi Public Networks.

A novel modification of IP networks integrated optimization method for heterogeneous networks, for example, the seamless Wi-Fi network serving simultaneously mobile users and wireless sensors, has been developed in this article. The mutual influence of signal reception, frequency-territorial planning, and routing procedures in heterogeneous networks have been analyzed in the case of simultaneous data transmission by both mobile users and wireless sensors. New principles for the listed procedures interaction and the basic functions for their describing are formulated. A novel modification of the integrated optimization method and its algorithm have been developed. The developed method's effectiveness has been analyzed for the IEEE 802.11ax network segment. Its result showed that the network load was decreased by an average of 20%, the data rate over the network as a whole increased for users and sensors by an average of 25% and 40%, respectively, and the sensors' lifetime increased by an average of 20% compared to the novel modification of the Collective Dynamic Routing method.

Novel Modification of the Collective Dynamic Routing Method for Sensors' Communication in Wi-Fi Public Networks.

The widespread use of the Internet of Things makes it relevant to use public IP networks for simultaneous access by both users and wireless sensors. To achieve this, a significant reduction in the subscriber devices' energy consumption is required. This paper analyzes the application features of the collective dynamic routing method both with and without the use of a robust method for estimating the channel data rate for sensors' communication in wireless public networks. Based on the analysis, a novel modification of the collective dynamic routing method has been developed that reduces the sensors' energy consumption while keeping a high data rate and short delivery time for users. An analysis of the network load, the total data transfer rate over the network, and the parameters affecting the sensors' energy consumption was carried out for a segment of a seamless IEEE 802.11ax network. The simulation demonstrated a high efficiency of using a novel modification of the collective dynamic routing method for access to users and wireless sensors.

A Design of a Time Synchronization Protocol Based on Dynamic Route and Forwarding Certification.

Time synchronization is a key technique in large-scale wireless sensor network applications. In order to tackle the problems of multi-hop synchronization error accumulation, clock frequency skew swinging, and network topology changes, a time synchronization protocol based on dynamic routing and forwarding certification (DRFC-TSP) is proposed in this paper. During the time synchronization process, a reference node with fewer synchronization hops and a more stable clock frequency is selected for every single hop, in order to obtain the best synchronization route. In this way, synchronization error accumulation can be restrained and the impact of clock frequency skew swinging on the time synchronization precision can be reduced. Furthermore, changes of the network topology can be well adapted by dynamic routing, in which the reference node is updated in every synchronization round. In the forwarding certification process, the status of nodes forwarding synchronous information outwards is authored by information exchange between neighboring nodes. Only synchronous information of the certificated nodes with a better performance can be forwarded. The network traffic can be decreased and the time synchronization precision can also be ensured, even with less energy consumption. Feasibility testing in large-scale wireless sensor networks is verified on NS2 simulation and more performances are evaluated on an embedded Linux platform.

An Improved Skin Lesion Classification Using a Hybrid Approach with Active Contour Snake Model and Lightweight Attention-Guided Capsule Networks.

Skin cancer is a prevalent type of malignancy on a global scale, and the early and accurate diagnosis of this condition is of utmost importance for the survival of patients. The clinical assessment of cutaneous lesions is a crucial aspect of medical practice, although it encounters several obstacles, such as prolonged waiting time and misinterpretation. The intricate nature of skin lesions, coupled with variations in appearance and texture, presents substantial barriers to accurate classification. As such, skilled clinicians often struggle to differentiate benign moles from early malignant tumors in skin images. Although deep learning-based approaches such as convolution neural networks have made significant improvements, their stability and generalization continue to experience difficulties, and their performance in accurately delineating lesion borders, capturing refined spatial connections among features, and using contextual information for classification is suboptimal. To address these limitations, we propose a novel approach for skin lesion classification that combines snake models of active contour (AC) segmentation, ResNet50 for feature extraction, and a capsule network with a fusion of lightweight attention mechanisms to attain the different feature channels and spatial regions within feature maps, enhance the feature discrimination, and improve accuracy. We employed the stochastic gradient descent (SGD) optimization algorithm to optimize the model's parameters. The proposed model is implemented on publicly available datasets, namely, HAM10000 and ISIC 2020. The experimental results showed that the proposed model achieved an accuracy of 98% and AUC-ROC of 97.3%, showcasing substantial potential in terms of effective model generalization compared to existing state-of-the-art (SOTA) approaches. These results highlight the potential for our approach to reshape automated dermatological diagnosis and provide a helpful tool for medical practitioners.

Performance evaluation of multi-channel wireless mesh networks with embedded systems.

Many commercial wireless mesh network (WMN) products are available in the marketplace with their own proprietary standards, but interoperability among the different vendors is not possible. Open source communities have their own WMN implementation in accordance with the IEEE 802.11s draft standard, Linux open80211s project and FreeBSD WMN implementation. While some studies have focused on the test bed of WMNs based on the open80211s project, none are based on the FreeBSD. In this paper, we built an embedded system using the FreeBSD WMN implementation that utilizes two channels and evaluated its performance. This implementation allows the legacy system to connect to the WMN independent of the type of platform and distributes the load between the two non-overlapping channels. One channel is used for the backhaul connection and the other one is used to connect to the stations to wireless mesh network. By using the power efficient 802.11 technology, this device can also be used as a gateway for the wireless sensor network (WSN).

Ensemble of deep capsule neural networks: an application to pediatric pneumonia prediction.

Pneumonia disease accounts for 15% of all deaths in children under the age of five and early detection of the disease significantly improves survival chances. In this work, we introduce a novel deep neural network model for evaluating pediatric pneumonia from chest radio-graph images. The proposed network is an ensemble of multiple candidate networks, each with interleaved convolutional and capsule layers. Individual networks are stitched together with dense layers and trained as a single model to minimize joint loss. The proposed approach is validated through extensive experimentation on the benchmark pneumonia dataset, and the results demonstrate that the model captures higher level abstractions as well as hidden low-level features from the input radio-graphic images. Our comparison studies reveal that the proposed model produces more generic predictions than existing approaches, with an accuracy of 94.84%. The proposed model produces better scores than the existing models and is extremely useful in assisting clinicians in pneumonia diagnosis.

Dynamic guided metric representation learning for multi-view clustering.

Multi-view clustering (MVC) is a mainstream task that aims to divide objects into meaningful groups from different perspectives. The quality of data representation is the key issue in MVC. A comprehensive meaningful data representation should be with the discriminant characteristics in a single view and the correlation of multiple views. Considering this, a novel framework called Dynamic Guided Metric Representation Learning for Multi-View Clustering (DGMRL-MVC) is proposed in this paper, which can cluster multi-view data in a learned latent discriminated embedding space. Specifically, in the framework, the data representation can be enhanced by multi-steps. Firstly, the class separability is enforced with Fisher Discriminant Analysis (FDA) within each single view, while the consistence among different views is enhanced based on Hilbert-Schmidt independence criteria (HSIC). Then, the 1st enhanced representation is obtained. In the second step, a dynamic routing mechanism is introduced, in which the location or direction information is added to fulfil the expression. After that, a generalized canonical correlation analysis (GCCA) model is used to get the final ultimate common discriminated representation. The learned fusion representation can substantially improve multi-view clustering performance. Experiments validated the effectiveness of the proposed method for clustering tasks.

CapsCovNet: A Modified Capsule Network to Diagnose COVID-19 From Multimodal Medical Imaging.

Since the end of 2019, novel coronavirus disease (COVID-19) has brought about a plethora of unforeseen changes to the world as we know it. Despite our ceaseless fight against it, COVID-19 has claimed millions of lives, and the death toll exacerbated due to its extremely contagious and fast-spreading nature. To control the spread of this highly contagious disease, a rapid and accurate diagnosis can play a very crucial part. Motivated by this context, a parallelly concatenated convolutional block-based capsule network is proposed in this article as an efficient tool to diagnose the COVID-19 patients from multimodal medical images. Concatenation of deep convolutional blocks of different filter sizes allows us to integrate discriminative spatial features by simultaneously changing the receptive field and enhances the scalability of the model. Moreover, concatenation of capsule layers strengthens the model to learn more complex representation by presenting the information in a fine to coarser manner. The proposed model is evaluated on three benchmark datasets, in which two of them are chest radiograph datasets and the rest is an ultrasound imaging dataset. The architecture that we have proposed through extensive analysis and reasoning achieved outstanding performance in COVID-19 detection task, which signifies the potentiality of the proposed model.

BF2 VHDR based dynamic routing with hydrodynamics for QoS development in WSN.

The Hydrodynamic characteristics has been considered for routing in Wireless Sensor Networks by various researchers and presented several methods. The flow and friction based routing approaches would produces sustainable results but would not improve the QoS. To improve the performance of F2VHDR (Flow -Flow-Friction-Velocity Based Hydro Dynamic Routing), this paper present BF2VHDR (Back Flow -Flow-Friction-Velocity Based Hydro Dynamic Routing) algorithm. The F2VHDR method misses the back flow of packets due to route failure or higher traffic conditions which affects the service performance. As a solution to this, the BF2VHDR algorithm is presented. The proposed BF2VHDR approach monitors the flow in both the sides of the route. The back flow occurs when it exist a route failure and higher traffic. Also, it may occur when the routing protocol of the other nodes would choose the reverse route as a best way to reach the same destination. Monitoring the back flow, general route flow, friction by traffic and velocity measures, the proposed method computes backward hydrology routing weight and forward hydrology routing weight. Using both the measures, the proposed method computes a route support weight for each route which has been used to perform route selection. The proposed approach improves the performance of throughput and increases the lifetime of the sensor nodes.

Investigating the transferring capability of capsule networks for text classification.

Text classification has been attracting increasing attention with the growth of textual data created on the Internet. Great progress has been made by deep neural networks for domains where a large amount of labeled training data is available. However, providing sufficient data is time-consuming and labor-intensive, establishing substantial obstacles for expanding the learned models to new domains or new tasks. In this paper, we investigate the transferring capability of capsule networks for text classification. Capsule networks are able to capture the intrinsic spatial part-whole relationship constituting domain invariant knowledge that bridges the knowledge gap between the source and target domains (or tasks). We propose an iterative adaptation strategy for cross-domain text classification, which adapts the source domain to the target domain. A fast training method with capsule compression and class-guided routing is designed to make the capsule network more efficient in computation for cross-domain text classification. We first conduct experiments to evaluate the performance of the capsule network on six benchmark datasets for generic text classification. The capsule networks outperform the compared models on 4 out of 6 datasets, suggesting the effectiveness of the capsule networks for text classification. More importantly, we demonstrate the transferring capability of the proposed cross-domain capsule network (TL-Capsule) by applying it to two transfer learning applications: single-label to multi-label text classification and cross-domain sentiment classification. The experimental results show that capsule networks consistently and substantially outperform the compared methods for both tasks. To the best of our knowledge, this is the first work that empirically investigates the transferring capability of capsule networks for text modeling.

Hetereogeneity in Neuronal Intrinsic Properties: A Possible Mechanism for Hub-Like Properties of the Rat Anterior Cingulate Cortex during Network Activity.

The anterior cingulate cortex (ACC) is vital for a range of brain functions requiring cognitive control and has highly divergent inputs and outputs, thus manifesting as a hub in connectomic analyses. Studies show diverse functional interactions within the ACC are associated with network oscillations in the ß (20-30 Hz) and γ (30-80 Hz) frequency range. Oscillations permit dynamic routing of information within cortex, a function that depends on bandpass filter-like behavior to selectively respond to specific inputs. However, a putative hub region such as ACC needs to be able to combine inputs from multiple sources rather than select a single input at the expense of others. To address this potential functional dichotomy, we modeled local ACC network dynamics in the rat in vitro. Modal peak oscillation frequencies in the ß- and γ-frequency band corresponded to GABAAergic synaptic kinetics as seen in other regions; however, the intrinsic properties of ACC principal neurons were highly diverse. Computational modeling predicted that this neuronal response diversity broadened the bandwidth for filtering rhythmic inputs and supported combination-rather than selection-of different frequencies within the canonical γ and ß electroencephalograph bands. These findings suggest that oscillating neuronal populations can support either response selection (routing) or combination, depending on the interplay between the kinetics of synaptic inhibition and the degree of heterogeneity of principal cell intrinsic conductances.