Search

In this contribution, we consider computed tomography (CT) as a diagnostic tool for identifying coronavirus disease 2019 (COVID-19) pneumonia. However, interpreting CT scans can be subjective, leading to interobserver variability and potential misdiagnosis. To address these challenges, a deep learning-based chest approach was developed to create a precise diagnostic tool for COVID-19 pneumonia and a personalized therapeutic strategy for individual patients. The study collected chest CT images from patients with different lung conditions, creating a diverse convolutional neural network (CNN) training material. Three different CNN-based models were tested for diagnostic purposes, with the output stating whether the patient was healthy or infected. The models facilitated selecting regions of interest (ROIs) and extracting the radiomic features from the input data, resulting in satisfactory results with core classification quality measures above the 50% threshold. For therapeutic purposes, a custom U-Net-based model was used to extract lung and infection masks from a provided CT slice. The percentage of the pathologically altered tissue was calculated, and the COVID-19 severity score was computed and then matched with an optimal therapeutic strategy. Overall, the models delivered high-quality results, representing a functioning deep learning-based application that could be advantageous as a doctor-friendly support tool. The use of deep learning techniques in medical imaging shows promising results, improving the accuracy and speed of diagnosis and treatment of not only COVID-19 but also many different diseases.

Artificial intelligence is one of the fastest-developing areas of science that covers a remarkably wide range of problems to be solved. It has found practical application in many areas of human activity, also in medicine. One of the directions of cooperation between computer science and medicine is to assist in diagnosing and proposing treatment methods with the use of IT tools. This study is the result of collaboration with the Children’s Memorial Health Institute in Warsaw, from where a database containing information about patients suffering from Bruton’s disease was made available. This is a rare disorder, difficult to detect in the first months of life. It is estimated that one in 70,000 to 90,000 children will develop Bruton’s disease. But even these few cases need detailed attention from doctors. Based on the data contained in the database, data mining was performed. During this process, knowledge was discovered that was presented in a way that is understandable to the user, in the form of decision trees. The best models obtained were used for the implementation of expert systems. Based on the data introduced by the user, the system conducts expertise and determines the severity of the course of the disease or the severity of the mutation. The CLIPS language was used for developing the expert system. Then, using this language, software was developed producing six expert systems. In the next step, experimental verification was performed, which confirmed the correctness of the developed systems.

This paper concerns the issues of investing in the stock market using artificial intelligence tools. It describes in detail the process of investing in capital markets using the combination of technical analysis and fuzzy logic. Therefore, a kind of an expert system used to predict future share prices has been developed. Its main goal is to support the investor in making decisions by suggesting the purchase of shares, their sale or refraining from acting. Firstly, the paper describes some technical analysis indicators that have been chosen during the literature review. Then, after selecting the appropriate indicators, the model of the fuzzy system has been designed. The created system has been divided into two main parts. The first one is used to calculate the values of the technical analysis indicators based on historical data. The second part is designed to generate a decision in the process of reasoning based on the results from the first part. To verify the decisions taken by the system, a series of tests based on historical data has been conducted. After comparing the results with the real price fluctuations the system decisions have been confirmed. Therefore, it has been concluded that the developed application can be a useful tool to invest in the stock market using technical analysis.

Classical logic is one of the most popular logic systems. However, it is quite an inaccurate model of reality, which was noticed already in ancient times. Two logics of belief are considered in this paper. In these logics, the agent forms his beliefs based on logical values of complex sentences. Both implemented logics are three-valued, in contrast to standard logic. For the needs of this paper, an application has been created which can be used to effectively solve satisfiability or tautology problems. To achieve this, the DPLL algorithm has been modified to handle three-valued logic. The paper also presents development and verification methods.

This paper concerns theorem proving in some extended logic. As the core part of this work the computer application implementing Wiśniewski’s logic has been developed. The article begins with the presentation of logic syntax and semantics. The next part is a description of the proving system which is based on resolution rule, which enables automatization of the process. The final part focuses on a description of steps used for proving theorems. The correctness of the algorithm has been verified by experiments.

Time is a concept which underlies all the contemporary civilization. Therefore, it was necessary to create mathematical tools that allow a precise way to describe the complex time dependencies. One such tool is temporal logic. Its definition, description and characteristics will be presented in this publication. Then the authors will conduct a discussion on the usefulness of this tool in context of creating storyline in computer games such as RPG genre.

Purpose The objective of this study was to evaluate thevitamin D concentration in patients with recurrent respiratory infections with or without immunoglobulin G, A or M (IgG, IgA, IgM) deficiency, and to find a correlation between the vitamin D concentration and the response to hepatitis B vaccination. Materials and method The study involved 730 patients with recurrent respiratory infections. The concentration of 25-hydroxyvitamin D (25(OH)D), immunoglobulins G, A and M, anti-HBs was determined. Results The tests showed that 11% of patients presented IgG levels below the age related reference values. Children with reduced IgG concentration were also found to have significantly lower vitamin D concentrations in comparison to children with normal IgG. Vitamin D deficiency was observed in schoolchildren between 7 and 18 years of age. No correlation was found between 25(OH)D concentration and Hbs antibody levels. Conclusions An investigation of a large group of patients who have recurrent infection found patients with IgG deficiency to whom special proceeding have to be performed: 1. Significantly lower vitamin D concentration observed in the group of children with IgG deficiency implicated in long-lasting monitoring of vitamin D level require adding to the practice guidelines for Central Europe 2013. 2. Intervention treatment with suitable doses of vitamin D to clarified metabolism of vitamin D has to be plan for children with IgG deficiency and significant lower vitamin D concentration.

Bioinformatics plays an important role in natural sciences. One of its branches – Computer-Aided Drug Design (CADD) – gives practical insights for designing and discovering of novel – better and safer – drugs. The CADD encompasses many different techniques like docking, virtual screening and quantitative structure-activity relationships (QSAR). The latter deals with building equations relating drug activities and their structures represented by variables called molecular descriptors. An important and promising type of such descriptors are Sinister-Rectus Chirality Measures (SRCM). However, the only so far available software for \(^{SR}CM\) calculation is very slow, and this impedes wider application of \(^{SR}CM\) by QSAR community. Therefore, an attempt to develop a novel algorithm for calculation of \(^{SR}CM\) (using Genetic Algorithm) and to implement it in an efficient and modern computer program was made. The result of these efforts is Chirmes. Performed tests have shown that Chirmes gives correct results of \(^{SR}CM\) calculations and performs way faster than the so far available software does. The paper describes first chemical and computational background behind the tackled problem. Then details of the implementation are presented, along with the test results and future prospects.

This paper presents the temporal logic inference engine developed in our university. It is an attempt to demonstrate implementation and practical application of temporal logic LNC developed in Cardinal Stefan Wyszynski University in Warsaw.1 The paper describes the fundamentals of LNC logic, architecture and implementation of inference engine. The practical application is shown by providing the solution for popular in Artificial Intelligence problem of Missionaries and Cannibals in terms of LNC logic. Both problem formulation and inference engine are described in details.

The real art of management it is important to make smart decisions, what in most of the cases is not a trivial task. Those decisions may lead to determination of production level, funds allocation for investments etc. Most of the parameters in decision-making process such as: interest rate, goods value or exchange rate may change. It is well know that these parameters in the decision-making are based on the data contained in datamarts or data warehouse. However, if the information derived from the processed data sets is the basis for the most important management decisions, it is required that the data is accurate, complete and current. In order to achieve high quality data and to gain from them measurable business benefits, data quality system should be used. The article describes the approach to the problem, shows the algorithms in details and their usage. Finally the test results are provide. Test results show the best algorithms (in terms of quality and quantity) for different parameters and data distribution.

Boosting algorithms constitute one of the essential tools in modern machine-learning, one of its primary applications being the improvement of classifier accuracy in supervised learning. Most widespread realization of boosting, known as AdaBoost, is based upon the concept of building a complex predictive model out of a group of simple base models. We present an approach for local assessment of base model accuracy and their improved weighting that captures inhomogeneity present in real-life datasets, in particular in those that contain geographic information. Conducted experiments show improvement in classification accuracy and F-scores of the modified algorithm, however more experimentation is required to confirm the exact scope of these improvements.

In this article, we are going to present the issue concerning the tautology for the three-valued logic of beliefs – LB3. The aim of the paper is to present a machine implementation of tautology analysis for the given logic. Due to the presented analysis, we can classify a given LB3 language formula as: tautology, contradiction, or a satisfiable function. We also present the method for decoding LB3 logic expressions, since machine implementation of any aspects related to logical systems require proper coding.

NoSQL is a new approach to data storage and manipulation. The aim of this paper is to gain more insight into NoSQL databases, as we are still in the early stages of understanding when to use them and how to use them in an appropriate way. In this submission descriptions of selected NoSQL databases are presented. Each of the databases is analysed with primary focus on its data model, data access, architecture and practical usage in real applications. Furthemore, the NoSQL databases are compared in fields of data references. The relational databases offer foreign keys, whereas NoSQL databases provide us with limited references. An intermediate model between graph theory and relational algebra which can address the problem should be created. Finally, the proposal of a new approach to the problem of inconsistent references in Big Data storage systems is introduced.

The purpose of this paper is to present the computer implementation of a system known as LC temporal logic [1]. Firstly, to become familiar with some theoretical issues, a short introduction to this logic is discussed. The algorithms allowing a deep analysis of the formulae of LC logic are considered. In particular we discuss how to determine if a formula is a tautology, contrtautology or it is satisfable. Next, we show how to find all valuations to satisfy the formula. Finally, we consider finding histories generated by the formula and transforming these histories into the state machine. Moreover, a description of the experiments that verify the implementation are briefly presented.

This paper concerns one of the fundamental issues of software engineering. We consider the concept of software metrics based on attributes describing each line of a computer program source code. Furthermore we present a design for a flexible, plugin based system which allows to define and use such metrics. To show the practical utility of the solution we implemented a basic version of the system and conducted experimental verification. Some of tests results and final conclusions are also provided.

This article shows the basic methods of electroencephalography EEG signal exploration. It contains information about data acquisition and different methods in which brain-computer interfaces can be made. The main focus of the paper is to find a way to determine the best set of parameters to detect movement of a hand in EEG signal. In the introduction there is also short introduction to EEG as well as fundamentals of support vector machine.

The paper addresses a new implementation of genetic programming by using molecular approach. Our method is based on dataflow techniques in DNA computing. After description of fundamental operations on DNA molecules and construction of logical functions the genetic programming method is introduced. We propose a way to handle these graph encoding molecules and which can be considered a genetic programming algorithm; a short discussion about experiments in implementing parts of this procedure is added.

Molecular computing is a new paradigm to perform calculations using nanotechnology. This paper presents the overall research direction from which molecular inference and expert systems are emerging. It introduces the subject matter and a general description of the problems involved. This includes selected methods of knowledge representation by DNA oligonucleotides, strategies of the inference mechanism, concept of the inference engine based on circular DNA molecules, particularly derived from plasmids, practical experience in DNA inference engine implementation, and discussion of the experimental results. The approach allows evaluating logical statements and drawing inferences for generating other statements via DNA computing. A series of experiments have been conducted to confirm practical utility of this approach. In these experiments, parameters of biochemical reactions were varied to determine truth/false recognition accuracy. In addition, we discuss the fundamental issues of inference engine and try ...

Our aim is to present the algebra of concepts in two formal languages. First, after introducing a primary relation between concepts, which is subsumption, we shall specify in a language that uses quantifiers, the Boolean algebra of general concepts. Next, we shall note down the same algebra in simplified non-quantifying language, in order to use it as basis for two specific implementations, i.e. to create the Boolean algebras of deontic concepts and axiological concepts.

The aim of this paper is to design and implement software application converting text of law acts documents from Microsoft Word file format to structural form of law acts in XML format. The structure of XML documents is consistent with the legislative rules and standards which are defined in XML Schema provided by Polish parliamentary units. This approach encompasses the following steps: analysis of software requirements, presentation of problem solving concept, design and implementation directions of the final application. Certain experiments and software testing were performed in order to verify the correctness of the solution. A brief summary and some comments were demonstrated in the final conclusion.

In this article there are presented a few methods of proving that can be adopted for epistemic logic. Those methods are used for proving sample thesis in a certain considered epistemic logic also described briefly in this work.

The article describes an attempt to implement abstract and concrete inherence calculi in the PowerLoom technology. Issues in the field of artificial intelligence, ontology and philosophy have been addressed. The inherence calculus is a type of a formal logic system. The PowerLoom technology consists of a knowledge representation language and an inference engine. Six inherence calculi have been implemented and an appropriate testing environment has been developed. The inherence calculus has been also extended by categorical properties and a theoretical discussion of ontological Boolean algebra has been conducted. Carried out experiments showed properties of the inherence calculi and also verified capabilities of PowerLoom to construct such logic systems. It occurred that expert system operational mode of PowerLoom outperforms its abilities to work as a mathematical theorem prover.

The aim of this paper is to design and implement information system containing large database dedicated to support neurological-psychiatric examinations focused on human brain after stroke. This approach encompasses the following steps: analysis of software requirements, presentation of the problem solving concept, design and implementation of the final information system. Certain experiments were performed in order to verify the correctness of the project ideas. The approach can be considered as an interdisciplinary venture. Elaboration of the system architecture, data model and the tools supporting medical examinations are provided. The achievement of the design goals is demonstrated in the final conclusion.

Adequate representation of data is crucial for modeling any type of data. To faithfully present and describe the relevant section of the world it is necessary to select the method that can easily be implemented on a computer system which will help in further description allowing reasoning. The main objective of this contribution is to present methods of knowledge representation using categorial approach. Next to identify the main advantages for computer implementation. Categorical aspect of knowledge representation is considered in semantic networks realisation. Such method borrows already known metaphysics properties for data modeling process. The potential topics of further development of categorical semantic networks implementations are also underlined.

The paper broaches topics of selected knowledge discovery algorithms. Different implementations have been verified on parallel platforms, including graphics accelerators using CUDA technology, multi-core microprocessors using OpenMP and many graphics accelerators. Results of investigations have been compared in terms of performance and scalability. Different types of data representation were also tested. The possibilities of both platforms, using the classification algorithms: the k-nearest neighbors, support vector machines and logistic regression are discussed.

The prototype model of classification assumes that categories are stored in human mind as abstracted summary representations formed in the process of experiencing specimens. Classification of new exemplars is based on their similarity to the abstracted prototype. From studies using Michael Posner’s dot-pattern recognition paradigm, we selected several empirical observations, like category size effect, category breadth effect or prototype-exemplar similarity effect, and tested them on artificial neural networks. In this work we show that the properties of human categorization process can be very well simulated and observed on artificial neural networks.

It has been recognized that ontologies are a semantic version of world wide web and can be found in knowledge-based systems. A recent time survey of this field also suggest that practical artificial intelligence systems may be motivated by this research. Especially strong artificial intelligence as well as concept of homo computer can also benefit from their use. The main objective of this contribution is to present and review already created ontologies and identify the main advantages which derive such approach for knowledge management systems. We would like to present what ontological engineering borrows from metaphysics and what a feedback it can provide to natural language processing, simulations and modelling. The potential topics of further development from philosophical point of view is also underlined.

The paper presents a digest of the research results in the domain of accelerator science and technology in Europe, shown during the annual meeting of the EuCARD – European Coordination of Accelerator Research and Development. The conference concerns building of the research infrastructure, including in this advanced photonic and electronic systems for servicing large high energy physics experiments. There are debated a few basic groups of such systems like: measurement – control networks of large geometrical extent, multichannel systems for large amounts of metrological data acquisition, precision photonic networks of reference time, frequency and phase distribution. Keywords: electronics and photonics for high energy physics experiments, free electron laser, distributed measurement and control systems, precise timing dist ribution systems of large space extent, ad vanced electronic systems, integration of hardware and software 1. INTRODUCTION The EUCARD 2011 Annual Conference on the development of the accelerator research infrastructure in Europe was held in Paris on 10-13 May. The venue of the meeting was the IN2P3 institute, a part of the CNRS. Around 150 accelerator researchers participated and 80 papers were presented. The conference covered the following subjects: material engineering – new materials for building accelerator sub-systems, building of research infrastructure for measurements of the mass and oscillations of neutrinos, muon electronics, upgrade of existing accelerators, HL-LHC, HE-LHC, building of new infrastructure of large scale, experiments of new physics, plasma wake field and laser accelerators.

In this paper we discuss fundamental issues of the conceptual knowledge representation. The primary objective of this contribution is to show some similarities between formal concept analysis and other mathematical structures. These similarities induce a number of questions to be broached in further interdisciplinary research.

The primary objective of this article is to find new effective methods of diagnosis of urgent liver transplantation after Amanita phalloides intoxication amongst pediatric patients. The research was carried out using a medical database of pediatric patients who suffered from acute liver failure after amatoxin consumption. After data preprocessing and attribute selection steps, a two-phase experiment was conducted, which incorporated a wide variety of data mining algorithms. The results deliver two equivalent classification models with simple decision structure and reasonable quality of surgery prediction.

The primary objective of this article is to find a new, more effective method of diagnosis of Crohn's disease. Having created the database on this disease we wanted to find the most suitable classification models. We used the algorithms with their implementations stored in R environment. Having carried out the investigations we have reached results interesting for clinical practice.

The paper describes an innovative approach of tensor calculations using parallel programming architecture. General use of a Graphic Processor Unit (GPU) in complicated and time-consuming operations on a large volume of data represented by tensor algebra is considered. Experimental verifications show practical utility of this approach. As a result, significant increase in speed of calculations compared with the classic sequential method have been demonstrated.

Epilepsy is a common nervous system disease often related to consciousness disturbances and muscular spasm which affects about 1% of the human population. Despite major technological advances done in medicine in the last years there was no sufficient progress towards overcoming it. Application of advanced statistical methods and computer image analysis offers the hope for accurate detection and later removal of an epileptiform focuses which are the cause of some types of epilepsy. The aim of this work was to create a computer system that would help to find and diagnose disorders of blood circulation in the brain This may be helpful for the diagnosis of the epileptic seizures onset in the brain.

The paper describes an innovative approach to discover new knowledge in non-alcoholic fatty liver disease (NAFLD). In order to determine the factors that may cause the disease a number of classification and attribute selection algorithms have been applied. Only those with the best classification results were chosen. Several interesting facts associated with this unclear disease have been discovered. All data mining computations were made in Weka environment.

The purpose of this paper is to figure out a new, better than so-far-known, evaluation method of diagnostic tests in Wilson disease. In order to find the most interesting classification models various data mining techniques were applied to real, suffering from Wilson disease, set of patients. It occurred that a combination of two classification algorithms with its implementations in Weka environment may significantly increase classification ability.

Age-related Macular Degeneration (AMD), according to epidemiological data, is a main reason of social blindness among elderly people in developed countries. There are two forms of AMD: dry and wet. The first one is of good prognosis with low possibility of serious visual deterioration, while the second one usually leads to quick and severe visual impairment. The aim of our investigations is to analyse results of so called real-life treatment of wet AMD. We analysed outcomes of our patients treated with intravitreal injections of anti-VEGF drugs: Lucentis (61 patients) and Avastin (78 patients). We analysed changes in visual acuity (functional effect) and central retinal thickness (anatomic effect). Both drugs occurred to be efficient in treatment of wet form of AMD, however results were more satisfying in patients with better baseline visual acuity. In our approach we used R environment - an integrated suite of software facilities for data analysis and graphics.

This paper presents three novel biquads derived by applying the concept of so-called basic elements for switched-capacitor (SC) circuits. These biquads are without continuous feedback and comprise two op-amps. They were derived assuming a transfer function denominator of the same form as the Fleischer-Laker type E circuit. It is shown that by a simple modification of the switching arrangement, i.e. by changing the type of the basic SC element, the effect of the amplifier finite gain on the performance of SC filters can be significantly reduced. the efficiency of this approach has been established by extensive simulation studies using the dedicated symbolic analysis program SCYMBAL. Design equations and comparison with already known biquads of this type are presented. Examples are provided to demonstrate that the proposed circuits require a low total capacitance while finite-gain effects are considerably reduced.

This paper describes an associative memory based on DNA strands practically build in laboratory. The method for suppressing DNA fragment amplification during polymerase chain reaction (PCR) was used. Such memory exhibits a number of advantages over Baum's associative molecular memory as well as traditional electronic implementations.

In this paper a new technique of sending data between molecular processors is presented. The molecular processor is a processing data unit. Its computation results have to be sent to other units in the form of addressed messages - tokens. Necessary experiments were performed. All operations were implemented in DNA. DNA processors and tokens were specially designed DNA strings. Results of experiments prove our assumptions.

In this paper we implement a new logic NAND gate using standard operations on DNA strands as well as digestion by the restriction nuclease class II. This concept despite some difficulties looks in general more elegant and can be utilized with fluorescent probes. Some experimental results demonstrating implementation of a single logic NAND gate are provided. The derived logic gates are proposed to be implemented on DNA chips.

DNA computing is a new paradigm to perform calculations using genetic engineering technology. The paper presents the overall research direction from which molecular inference and expert systems are emerging. It provides an introduction to the subject matter and a general description of the problems involved. This includes selected methods of knowledge representation by DNA strands, strategies of the inference mechanism, concept of the inference engine based on circular DNA molecules, particularly derived from plasmids, practical experience in DNA inference engine implementation, and discussion of the experimental results. The approach allows evaluating logical statements and drawing inferences for generating other statements via DNA computing.

A novel algorithm based on DNA computing for adding binary integer numbers is presented. It requires the unique representation of bits placed in test tubes treated as registers. Amplification step used for the carry operation allows one, in theory, to add numbers with the same quantity of elementary operations, regardless of the number of bits used for representation. New notation proposed in the paper allows for efficient and abstract description of the technical operations on DNA.

Self-assembly of DNA is considered a fundamental operation in realization of molecular logic circuits. We propose a new approach to implementation of data flow logical operations based on manipulating DNA strands. In our method the logic gates, input, and output signals are represented by DNA molecules. Each logical operation is carried out as soon as the operands are ready. This technique employs standard operations of genetic engineering including radioactive labeling. To check practical utility of the method a series of genetic engineering experiments have been performed. The obtained results confirm interesting properties of the DNA-based molecular data flow logic gates. This technique may be utilized in massively parallel computers.

Inference process plays an important role in the realisation of expert systems. In this paper it is shown that chemical reactions may by used to perform molecular inference according to the algorithm of forward chaining. This method is accomplished by an adequate interpretation of inorganic chemical compounds and unidirectional reactions. In our approach premise clauses are represented by the reactants while conclusion clauses are represented by the products of reaction. Different inorganic compounds and reactions have been discussed with respect to their utility for the molecular inference. Special attention is focused on qualitative chemistry and a number of reactions has been taken into account. Experimental results demonstrating application of these reactions in expert systems are provided.

The problem of temporal credit assignment in reinforcement learning is typically solved using algorithms based on the methods of temporal differences TD(λ). Of those, Q-learning is currently best understood and most widely used. Using TD-based algorithms with λ > 0 often allows one to speed up the propagation of credit significantly, but it involves certain implementational problems. The traditional implementation of TD(λ > 0) based on eligibility traces suffers from lack of generality and computational inefficiency. The TTD ( Truncated Temporal Differences) procedure is a simple TD(λ) approximation technique that appears to overcome these drawbacks of eligibility traces. The paper outlines this technique, discusses its computational efficiency advantages, and presents experimental studies with the combination of TTD and Q-learning in deterministic and stochastic environments. These experiments show that TTD makes it possible to obtain a significant learning speedup without reducing reliability at essentially the same computational cost as usual TD(0) learning. We conclude that the TTD procedure is probably the most promising way of using TD methods for reinforcement learning, especially for tasks with large state spaces and a hard temporal credit assignment problem.