Wyniki wyszukiwania - BazTech

1

Surgical phase classification and operative skill assessment through spatial context aware CNNs and time-invariant feature extracting autoencoders

Pradeep Chakka Sai, Sinha Neelam

Biocybernetics and Biomedical Engineering

|

2023

|

Vol. 43, no. 4

700--724

EN

Automated surgical video analysis promises improved healthcare. We propose novel spatial context aware combined loss function for end-to-end Encoder-Decoder training for Surgical Phase Classification (SPC) on laparoscopic cholecystectomy (LC) videos. Proposed loss function leverages on fine-grained class activation maps obtained from fused multi-layer Layer-CAM for supervised learning of SPC, obtaining improved Layer-CAM explanations. Post classification, we introduce graph theory to incorporate known hierarchies of surgical phases. We report peak SPC accuracy of 96.16%, precision of 94.08% and recall of 90.02% on public dataset Cholec80, with 7 phases. Our proposed method utilizes just 73.5% of parameters as against existing state-of-the-art methodology, achieving improvement of 0.5% in accuracy, 1.76% in precision with comparable recall, with an order less standard deviation. We also propose DNN based surgical skill assessment methodology. This approach utilizes surgical phase prediction scores from the final fully-connected layer of spatial-context aware classifier to form multi-channel temporal signal of surgical phases. Time-invariant representation is obtained from this temporal signal through time- and frequency-domain analyses. Autoencoder based time-invariant features are utilized for reconstruction and identification of prominent peaks in dissimilarity curves. We devise a surgical skill measure (SSM) based on spatial-context aware temporal-prominence-of-peaks curve. SSM values are expected to be high when executed skillfully, aligning with expert assessed GOALS metric. We illustrate this trend on Cholec80 and m2cai16-tool datasets, in comparison with GOALS metric. Concurrence in the trend of SSM with respect to GOALS metric is obtained on these test videos, making it a promising step towards automated surgical skill assessment.

2

Tendencies of Chinese subways’ spatial growth in 2000-2020

Panov Roman

Prace Komisji Geografii Komunikacji PTG

|

2022

|

nr 25(2)

42--51

EN

One of the main transportation problems of biggest modern cities is the excessively high load on ground transport, which is why the development of subway networks is of particular importance. This article analyzes the development of the spatial structure of subway networks in China. Currently China shows an intensive growth of existing networks and massive openings of new networks, which makes it the most suitable object for studying the evolution of subway networks. The methodology developed by K. Kansky and S. A. Tarkhov was used as the theoretical basis of this study. The study was conducted by analyzing the dynamics of the main quantitative and topomorphological indicators of subway networks during their passage through the stages of spatial evolution. The following indicators were used: the number of subways, the total length of the network, the number of cycles in the network, the number of topological layers and the number of cycles in each of them, the number of branching tiers, the area of topological layers and their share in the cyclic core of the network, the distribution of the length between the elements of the network structure, average cycle size, topological limit, cyclization index and circuity index. We identified the patterns for passing the stages of evolutionary development by the networks of Chinese subways; also, we found common features that define the “Chinese” type of subway, we identified a new subtype of networks.

3

A note on possible density and diameter of countere xamples to the Seymour’s second neighborhood conjecture

Zelenskiy Oleksiy, Darmosiuk Valentyna, Nalivayko Illia

Opuscula Mathematica

|

2021

|

Vol. 41, no. 4

601--605

EN

Seymour’s second neighborhood conjecture states that every simple digraph without loops or 2-cycles contains a vertex whose second neighborhood is at least as large as its first. In this paper we show, that from falsity of Seymour’s second neighborhood conjecture it follows that there exist strongly-connected counterexamples with both low and high density (dense and sparse graph). Moreover, we show that if there is a counterexample to conjecture, then it is possible to construct counterexample with any diameter k ≥ 3

4

Metoda upraszczania struktur sieci gazowych

Osiadacz Andrzej J., Strzałkowski Krzysztof, Kotyński Łukasz

Gaz, Woda i Technika Sanitarna

|

2021

|

Nr 10

2--6

PL

W artykule omówiono metodę upraszczania struktury sieci gazowej. Metoda upraszcza strukturę eliminując część elementów sieci zachowując przy tym parametry hydrauliczne sieci rzeczywistej. Weryfikacji metody dokonano upraszczając struktury wielu sieci rzeczywistych.

EN

The paper presents a method for the reduction of gas networks structure. The method reduces a network structure equivalent to the original one, but which contains fewer components. The method was validated by simplifying many gas networks.

5

An ant algorithm for the maximum number of 3-cliques in 3-partite graphs

Schiff Krzysztof

Control and Cybernetics

|

2021

|

Vol. 50, No. 2

347--358

EN

The problem of finding the maximum number of d- vertices cliques (d = 3) in d-partite graph (d = 3) when graph density q is lower than 1 is an important problem in combinatorial optimization and it is one of many NP-complete problems. For this problem a meta-heuristic algorithm has been developed, namely an ant colony optimization algorithm. In this paper a new development of this ant algorithm and experimental results are presented. The problem of finding the maximum number of 3-vertices cliques can be encountered in computer image analysis, computer vision applications, automation and robotic vision systems. The optimal solution of this problem boils down to finding a set of 3-vertices cliques in a 3-partite graph and this set should have cardinality as high as possible. The elaborated ant colony algorithm can be easily modified for d-dimensional problems, that is for finding the maximum number of d-vertices cliques in a d-partite graph.

6

Multi-sourced power system restoration strategy based on modified Prim’s algorithm

Łukaszewski Artur, Nogal Łukasz

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2021

|

Vol. 69, nr 5

art. no. e137942

EN

Self-healing grids are one of the most developing concepts applied in electrical engineering. Each restoration strategy requires advanced algorithms responsible for the creation of local power systems. Multi-agent automation solutions dedicated for smart grids are mostly based on Prim’s algorithm. Graph theory in that field also leaves many problems unsolved. This paper is focused on a variation of Prim’s algorithm utility for a multi-sourced power system topology. The logic described in the paper is a novel concept combined with a proposal of a multi-parametrized weight calculation formula representing transmission features of energy delivered to loads present in a considered grid. The weight is expressed as the combination of three elements: real power, reactive power, and real power losses. The proposal of a novel algorithm was verified in a simulation model of a power system. The new restoration logic was compared with the proposal of the strategy presented in other recently published articles. The novel concept of restoration strategy dedicated to multi-sourced power systems was verified positively by simulations. The proposed solution proved its usefulness and applicability.

7

Rotation Invariance in Graph Convolutional Networks

Mac Nguyen Anh, Nguyen Hung Son

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

81--90

EN

Convolution filters in deep convolutional networks display rotation variant behavior. While learned invariant behavior can be partially achieved, this paper shows that current methods of utilizing rotation variant features can be improved by proposing a grid-based graph convolutional network. We demonstrate that Grid-GCN heavily outperforms existing models on rotated images, and through a set of ablation studies, we show how the performance of Grid-GCN implies that there exist more performant methods to utilize fundamentally rotation variant features and we conclude that the inherit nature of spectral graph convolutions is able to learn invariant behavior.

8

Discovering Communities in Networks: A Linear Programming Approach Using Max-Min Modularity

Ferdowsi Arman, Khanteymoori Alireza

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

329–-335

EN

Community detection is a fundamental challenge in network science and graph theory that aims to reveal nodes' structures. ‎While most methods consider Modularity as a community quality measure‎, ‎Max-Min Modularity improves the accuracy of the measure by penalizing the Modularity quantity when unrelated nodes are in the same community‎. ‎In this paper‎, ‎we propose a community detection approach based on linear programming using Max-Min Modularity‎. ‎The experimental results show that our algorithm has a better performance than the previously known algorithms on some well-known instances‎.

9

Combinatorial Etude

Stoenchev Miroslav, Todorov Venelin

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

231--234

EN

The purpose of this article is to consider a special class of combinatorial problems, the so called Prouhet-Tarry Escot problem, solution of which is realized by constructing finite sequences of ±1. For example, for fixed p∈N, is well known the existence of np∈N with the property: any set of np consecutive integers can be divided into 2 sets, with equal sums of its p[th] powers. The considered property remains valid also for sets of finite arithmetic progressions of complex numbers.

10

An Efficient Connected Swarm Deployment via Deep Learning

Danilchenko Kiril, Segal Michael

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

1--7

EN

In this paper, an unmanned aerial vehicles (UAVs) deployment framework based on machine learning is studied. It aims to maximize the sum of the weights of the ground users covered by UAVs while UAVs forming a connected communication graph. We focus on the case where the number of UAVs is not necessarily enough to cover all ground users. We develop an UAV Deployment Deep Neural network (mod) as a UAV's deployment deep network method. Simulation results demonstrate that mod can serve as a computationally inexpensive replacement for traditionally expensive optimization algorithms in real-time tasks and outperform the state-of-the-art traditional algorithms.

11

An Efficient Approach Towards the Generation and Analysis of Interoperable Clinical Data in a Knowledge Graph

Dörpinghaus Jens, Weil Vera, Schaaf Sebastian, Hübenthal Tobias

Annals of Computer Science and Information Systems

|

2021

|

Vol. 25

59--68

EN

Knowledge graphs have been shown to play an important role in recent knowledge mining and discovery, for example in the field of life sciences or bioinformatics. Contextual information is widely used for NLP and knowledge discovery in life sciences since it highly influences the exact meaning of natural language and also queries for data. The contributions of this paper are (1) an efficient approach towards interoperable data, (2) a runtime analysis of 14 real world use cases represented by graph queries and (3) a unique view on clinical data and its application combining methods of algorithmic optimisation, graph theory and data science.

12

The reactive power and voltage control management strategy based on virtual reactance cloud control

Zhang Wei Min, Zhang Yan Xia

Archives of Electrical Engineering

|

2020

|

Vol. 69, nr 4

921--936

EN

The paper aims at the higher reactive power management complexity caused by the access of distributed power, and the problem such as large data exchange capacity, low accuracy of reactive power distribution, a slow convergence rate, and so on, may appear when the controlled objects are large. This paper proposes a reactive power and voltage control management strategy based on virtual reactance cloud control. The coupling between active power and reactive power in the system is effectively eliminated through the virtual reactance. At the same time, huge amounts of data are treated to parallel processing by using the cloud computing model parallel distributed processing, realize the uncertainty transformation between qualitative concept and quantitative value. The power distribution matrix is formed according to graph theory, and the accurate allocation of reactive power is realized by applying the cloud control model. Finally, the validity and rationality of this method are verified by testing a practical node system through simulation.

13

Optimization of Retrieval Algorithms on Large Scale Knowledge Graphs

Dörpinghaus Jens, Stefan Andreas

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

227--236

EN

Knowledge graphs have been shown to play an important role in recent knowledge mining and discovery, for example in the field of life sciences or bioinformatics. Although a lot of research has been done on the field of query optimization, query transformation and of course in storing and retrieving large scale knowledge graphs the field of algorithmic optimization is still a major challenge and a vital factor in using graph databases. Few researchers have addressed the problem of optimizing algorithms on large scale labeled property graphs. Here, we present two optimization approaches and compare them with a naive approach of directly querying the graph database. The aim of our work is to determine limiting factors of graph databases like Neo4j and we describe a novel solution to tackle these challenges. For this, we suggest a classification schema to differ between the complexity of a problem on a graph database. We evaluate our optimization approaches on a test system containing a knowledge graph derived biomedical publication data enriched with text mining data. This dense graph has more than 71M nodes and 850M relationships. The results are very encouraging and -- depending on the problem -- we were able to show a speedup of a factor between 44 and 3839.

14

On Finding the Optimal Tree of a Complete Weighted Graph

Hosseini Seyed Soheil, Wormald Nick, Tian Tianhai

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

271--275

EN

We want to find a tree where the path length between any two vertices on this tree is as close as possible to their corresponding distance in the complete weighted graph of vertices upon which the tree is built. We use the residual sum of squares as the optimality criterion to formulate this problem, and use the Cholesky decomposition to solve the system of linear equations to optimise weights of a given tree. We also use two metaheuristics, namely Simulated Annealing (SA) and Iterated Local Search (ILS) to optimise the tree structure. Our results suggest that SA and ILS both perform well at finding the optimal tree structure when the dispersion of distances in the complete graph is large. However, when the dispersion of distances is small, only ILS has a solid performance.

15

MD-jeep: a New Release for Discretizable Distance Geometry Problems with Interval Data

Mucherino Antonio, Gonçalves Douglas S., Liberti Leo, Lin Jung-Hsin, Lavor Carlile, Maculan Nelson

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

289--294

EN

With the most recent releases of MD-JEEP, new relevant features have been included to our software tool. MD-JEEP solves instances of the class of Discretizable Distance Geometry Problems (DDGPs), which ask to find possible realizations, in a Euclidean space, of a simple weighted undirected graph for which distance constraints between vertices are given, and for which a discretization of the search space can be supplied. Since its version 0.3.0, MD-JEEP is able to deal with instances containing interval data. We focus in this short paper on the most recent release MD-JEEP 0.3.2: among the new implemented features, we will focus our attention on three features: (i) an improved procedure for the generation and update of the boxes used in the coarse-grained representation (necessary to deal with instances containing interval data); (ii) a new procedure for the selection of the so-called discretization vertices (necessary to perform the discretization of the search space); (iii) the implementation of a general parser which allows the user to easily load DDGP instances in a given specified format. The source code of MD-JEEP 0.3.2 is available on GitHub, where the reader can find all additional details about the implementation of such new features, as well as verify the effectiveness of such features by comparing MD- JEEP 0.3.2 with its previous releases.

16

Knowledge Detection and Discovery using Semantic Graph Embeddings on Large Knowledge Graphs generated on Text Mining Results

Dörpinghaus Jens, Jacobs Marc

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

169--178

EN

Knowledge graphs play a central role in big data integration, especially for connecting data from different domains. Bringing unstructured texts, e.g. from scientific literature, into a structured, comparable format is one of the key assets. Here, we use knowledge graphs in the biomedical domain working together with text mining based document data for knowledge extraction and retrieval from text and natural language structures. For example cause and effect models, can potentially facilitate clinical decision making or help to drive research towards precision medicine. However, the power of knowledge graphs critically depends on context information. Here we provide a novel semantic approach towards a context enriched biomedical knowledge graph utilizing data integration with linked data applied to language technologies and text mining. This graph concept can be used for graph embedding applied in different approaches, e.g with focus on topic detection, document clustering and knowledge discovery. We discuss algorithmic approaches to tackle these challenges and show results for several applications like search query finding and knowledge discovery. The presented remarkable approaches lead to valuable results on large knowledge graphs.

17

Intuitionistic Fuzzy Hamiltonian Cycle by Index Matrices

Traneva Velichka, Tranev Stoyan

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

345--348

EN

In this paper, the algorithm for finding a Hamiltonian cycle in an intuitionistic fuzzy graph (IFG) is proposed, based on the theories of intuitionistic fuzzy sets (IFSs) and of index matrices (IMs). The aim of the paper is to extend the algorithm to find a fuzzy Hamiltonian cycle (FHC) in an IFG to the intuitionistic fuzzy (IFHC) using the IFSs and IMs concepts. An intuitionistic fuzzy graph example about network of Wizz air airlines is modeled by the extended IM to illustrate the proposed algorithm. In the paper also are introduced for the first time three index-type operations over IMs.

18

How well a multi-model database performs against its single-model variants: Benchmarking OrientDB with Neo4j and MongoDB

Macak Martin, Stovcik Matus, Buhnova Barbora, Merjavy Michal

Annals of Computer Science and Information Systems

|

2020

|

Vol. 21

463--470

EN

Digitalization is currently the key factor for progress, with a rising need for storing, collecting, and processing large amounts of data. In this context, NoSQL databases have become a popular storage solution, each specialized on a specific type of data. Next to that, the multi-model approach is designed to combine benefits from different types of databases, supporting several models for data. Despite its versatility, a multi-model database might not always be the best option, due to the risk of worse performance comparing to the single-model variants. It is hence crucial for software engineers to have access to benchmarks comparing the performance of multi-model and single-model variants. Moreover, in the current Big Data era, it is important to have cluster infrastructure considered within the benchmarks. In this paper, we aim to examine how the multi-model approach performs compared to its single-model variants. To this end, we compare the OrientDB multi-model database with the Neo4j graph database and the MongoDB document store. We do so in the cluster setup, to enhance state of the art in database benchmarks, which is not yet giving much insight into cluster-operating database performance.

19

Liczba chromatyczna Thue'go

Pawlik Bartłomiej

MINUT Matematyka i Informatyka na Uczelniach Technicznych

|

2019

|

Nr 1

36-41

PL

W artykule przedstawione jest pojęcie ciągu niepowtarzalnego wraz z klasycznym twierdzeniem Axela Thue'go. Tematyka ciągów niepowtarzalnych w połączenia z pewnymi aspektami teorii grafów doprowadziła do powstania pojęcia tzw. liczby chromatycznej Thue'go grafu. Ma ona kilka nieoczywistych własności, które zostały zaprezentowane w drugiej części tekstu.

20

Reducing pollution levels generated by short sea shipping : use of Bayesian networks to analyse the utilization of liquefied natural gas as an alternative fuel

Molina Serrano Beatriz, González Cancelas Nicoleta, Soler Flores Francisco

Journal of KONES

|

2019

|

Vol. 26, No. 1

147--158

EN

Pollution adjacent to the continent's shores has increased in the last decades, so it has been necessary to establish an energy policy to improve environmental conditions. One of the proposed solution was the search of alternative fuels to the commonly used in Short Sea Shipping to reduce pollution levels in Europe. Studies and researches show that liquefied natural gas could meet the European Union environmental requirements. Even environmental benefits are important; currently there is not significant number of vessels using it as fuel. Moreover, main target of this article is exposing result of a research in which a methodology to establish the most relevant variables in the decision to implement liquefied natural gas in Short Sea Shipping has been development using data mining. A Bayesian network was constructed because this kind of network allows to get graphically the relationships between variables and to determine posteriori values that quantify their contributions to decision-making. Bayesian model has been done using data from some European countries (European Union, Norway and Iceland) and database was generated by 35 variables classified in 5 categories. Main obtained conclusion in this analysis is that variables of transport and international trade and economy and finance are the most relevant in the decision-making process when implementing liquefied natural gas. Even more, it can be stablish that capacity of liquefied natural gas regasification terminals under construction and modal distribution of water cargo transportation continental as the most decisive variables because they are the root nodes in the obtained network.