Papers by Egidijus Kazanavicius
The Paradigm of an Explainable Artificial Intelligence (XAI) and Data Science (DS)-Based Decision Support System (DSS)
Studies in computational intelligence, 2023

Sustainability
There are different approaches in different areas of what the risk is. ISO 31000 risk management ... more There are different approaches in different areas of what the risk is. ISO 31000 risk management standards describe risk as the effect of uncertainty on objectives. Many existing risk assessment procedures are based on the assumption that risk is the amount of any damage or loss multiplied by the probability of an event that could cause the damage. We are proposing a new risk approach, based on Hillson’s positive risk philosophy, that risk is not just a threat but also a composition of new opportunities, efforts that need to be put in, and uncertainty. For this approach, we composed a risk formula and a methodology based on that formula. A prototypical software tool was developed, and an experiment was performed using this tool to evaluate the risk of several interconnected projects and validate the developed risk assessment methodology. It should be mentioned that, in the methodology, the decision-making process is performed traceably; therefore, it can be stated that it has explai...
International Journal of Scientific and Research Publications (IJSRP), 2021
This paper suggests to extend the earlier proposed XAI-based medical diagnostic decision support ... more This paper suggests to extend the earlier proposed XAI-based medical diagnostic decision support system with a new functionality for the treatment and care elderly patients. The proposed extension is considered as a partial case of the universalized fuzzy SWOT maps network. The main idea was realized in a module called as LEVERAGE and was successfully simulated. The reliable results presented in this paper have confirmed a vitality of the idea. This paper also advocates the breakthrough in believes that the theories about diseases are good ones only if they offer treatment as well. The paper concludes with a list of necessary future research.
International Journal of Scientific and Research Publications (IJSRP), 2020
The paper presents a model of the geriatric medical diagnostic support system (MDSS) developed in... more The paper presents a model of the geriatric medical diagnostic support system (MDSS) developed in the context of artificial intelligence (AI) using elements of the explainable artificial intelligence (XAI) paradigm, such as computing with words (CWW), fuzzy SWOT analysis maps (FSM), possibility of verbal evaluation of sixteen symptoms, and estimates of the four interacting syndromes and their verbal interpretations. The functional organization of the model is presented mathematically using a vector description method and fuzzy reasoning operations. The viability and proper functioning of the implemented software model are confirmed by a pilot advisory practice in a geriatrics clinic. The paper concludes with a list of future further research.

Medicina, 2019
Background and objective: The successful adoption of technology is becoming increasingly importan... more Background and objective: The successful adoption of technology is becoming increasingly important to functional independence and successful ageing in place. A better understanding of technology usage amongst older people may help to direct future interventions aimed at improving their healthcare. We aimed to obtain the first data regarding technology use, including gerontechnologies, represented by fall detectors, from older adults in Lithuania. Material and methods: The research was carried out in the framework of the project Smart Gerontechnology for Healthy Ageing, which involved assessing the use of technologies and the readiness to use gerontechnologies, as represented by fall detectors. A total of 375 individuals that were more than 60 years of age were enrolled in the study. The self-reporting questionnaires were completed by geriatric in-patients, hospitalized in the geriatric department, and also by community-dwelling older adults. Results: Geriatric in-patients’ use of co...
Computational intelligence and neuroscience, 2018
Cryptographic frameworks depend on key sharing for ensuring security of data. While the keys in c... more Cryptographic frameworks depend on key sharing for ensuring security of data. While the keys in cryptographic frameworks must be correctly reproducible and not unequivocally connected to the identity of a user, in biometric frameworks this is different. Joining cryptography techniques with biometrics can solve these issues. We present a biometric authentication method based on the discrete logarithm problem and Bose-Chaudhuri-Hocquenghem (BCH) codes, perform its security analysis, and demonstrate its security characteristics. We evaluate a biometric cryptosystem using our own dataset of electroencephalography (EEG) data collected from 42 subjects. The experimental results show that the described biometric user authentication system is effective, achieving an Equal Error Rate (ERR) of 0.024.
Aukštesnių eilių statistika grįsto balso detektavimo algoritmo sudarymas ir tyrimas
This article proposes a new conception of the multi-platform software and fi rmware development t... more This article proposes a new conception of the multi-platform software and fi rmware development tool. This conception is aimed at improving the quality as well as increasing the development productivity in multi-platform systems. The Model Driven Architecture (MDA) offers a more effi cient software engineering process by raising the level of abstraction. A semi-formal Unifi ed Modeling Language (UML) and its extension Systems Modeling Language (SysML) can bring software developers to a more formal model description, especially for real-time applications, enhancing possibilities to model the dynamics of the software model and decomposition of the physical architecture. The illustration of this tool conception is based on creation of a VoIP (Voice over IP) system.
Information Technology And Control, 2015
This work presents the solution based on the augmenting sequence of linear programming problems (... more This work presents the solution based on the augmenting sequence of linear programming problems (LPP) as a tool for intellectualizing home environment. The proposed solution empowers the intelligent decision making procedure which can be applied to various intelligent control applications. The augmenting self-training procedure based on LPP approach is presented as well, which allows making reasonable decisions having only limited data about the controlled environment. The method permits retraining the decision making system when new data is available. As a proof of concept, this solution is applied to intelligent light control application. The obtained simulation results show the method's capability in making reasonable decisions according to users preferences.

2020 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2020
The classical SWOT methodology and many of the tools based on it used so far are very static, use... more The classical SWOT methodology and many of the tools based on it used so far are very static, used for one stable project and lacking dynamics [1]. This paper proposes the idea of combining several SWOT analyses enriched with computing with words (CWW) paradigm into a single network. In this network, individual analysis of the situation is treated as the node. The whole structure is based on fuzzy cognitive maps (FCM) that have forward and backward chaining, so it is called fuzzy SWOT maps. Fuzzy SWOT maps methodology newly introduces the dynamics that projects are interacting, what exists in a real dynamic environment. The whole fuzzy SWOT maps network structure has explainable artificial intelligence (XAI) traits because each node in this network is a "white box"—all the reasoning chain can be tracked and checked why a particular decision has been made, which increases explainability by being able to check the rules to determine why a particular decision was made or why ...

This paper presents a real time pulse wave identification method based on a multi-agent system. M... more This paper presents a real time pulse wave identification method based on a multi-agent system. Medical IoT (Internet-of-things) systems are rather restricted in terms of resource consumption and processing time, especially in emergency situations, thus decision making processes must comply real-time requirements. This research mainly focuses on assistive healthcare domain. The environment is treated as IoT network with proactive sensing, actuation and decision support capabilities. Multi-agent systems are proposed as a useful binding technology, encapsulating IoT and healthcare domains as well as serving as an engineering technique for this class of proactive timecritical systems. As an experiment, the proposed agent-based IoT healthcare systems’ development method was applied for pulse wave identification. The results have proved that agent technology is beneficial in IoT healthcare domain in terms of processing time, capable of achieving real time decision making in emergent situ...

Blood Pressure Estimation According to Photoplethysmographic Signal
This paper proposes a method for indirect systolic blood pressure estimation based on photoplethy... more This paper proposes a method for indirect systolic blood pressure estimation based on photoplethysmographic signal steepness. The proposed method was evaluated on 10 healthy subjects, aged 25±3 yrs., including 8 males and 2 females. The photoplethysmographic signal was obtained using a photoplethysmography device that is put on a finger, while reference blood pressure was measured using a semi-automatic OMRON blood pressure monitor. A significant 0.8550.025 (p < 0.001) correlation between the photoplethysmographic signal steepness and systolic blood pressure was discovered. The estimated systolic blood pressure obtained using the proposed method differed from the reference blood pressure by 3.8±4.2mmHg. The obtained results meet the AAMI (Association for the Advancement of Medical Instrumentation) requirements (5 ± 8mmHg). The results demonstrate that the photoplethysmographic signal wavefront can be applied successfully in wearable devices that can be used for 24 hour registrat...
Process network is a computation model in which many concurrent processes communicate through unb... more Process network is a computation model in which many concurrent processes communicate through unbounded FIFO buffer and can be executed simultaneously. In real time digital signal processing applications execution time is infinite. However, failures of implementation hardware can occur. In our work, dynamic run-time reconfiguration is introduced into process network which ensures error handling, avoiding deadlocks, continuous and on-time result delivery. After dynamic reconfiguration, network execution results may become non deterministic, but this helps avoiding critical termination of network execution. In this paper, we present a description of possible failures of network execution and discuss the means for avoiding these failures.

Hybrid Explainable Smart House Control System
2021 3rd International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA)
This paper describes hybrid method for smart house control system, where control is made by train... more This paper describes hybrid method for smart house control system, where control is made by training neural network with user habit data and then applying fuzzy rule set and computing with words engine to provide user with explanation why control change to light, heating or ventilation was made. For neural networks two different approaches were tested: classical support vector machine and newly emerged ML.Net framework. Both methods could correctly predict user desired controls with ~99% accuracy, but ML.Net was recommended for further use because it was less selective for the data format, on the other hand SVM required data separation for binary classification thus increasing required number of SVM machines if control areas would expand. Given verbal explanations accurately described current house control situation to the users, thus increasing confidence level for Explainable Artificial Intelligence.

Mathematics and Computer Science
The paper advocates a new concept for risk control that makes up one organic closed loop feedback... more The paper advocates a new concept for risk control that makes up one organic closed loop feedback system, with the following stages: 1) the evaluation of the positive and negative features of situation under investigation through strengths, weaknesses, opportunities, and threats (SWOT) analysis, 2) the determination of the level of fuzzy risk concealed in this situation (using RISK evaluation), and 3) the proposal of leverage, recommendations, or actions (through LEVERAGE aggregation) enabling the improvement of target performance. Useful fundamental approaches, definitions, and particularities of this concept concerning SWOT, RISK and LEVERAGES are examined, and for the first time the network type called here the fuzzy SWOT map (FSM) is introduced. This newly proposed instrument appeared as a result of a natural extension of fuzzy cognitive maps paradigm enhanced by dynamic computing with words (CWW) elements and possibilities to use the explainable artificial intelligence (XAI) in the form of fuzzy inference rules. The concept serves for development of functional organization of control systems of complex and dynamically interacting projects or situations and for implementation of adequate set of tools satisfying the concrete system's requirements. The results of conceptual modeling and the confirmation of the vitality of the approach are presented based on the simplified example of a risk-control system case covering three interacting projects in a complex environment of city development.

Informatica
Petri net variants are widely used as a real time systems modeling technique. Recently, UML activ... more Petri net variants are widely used as a real time systems modeling technique. Recently, UML activity diagrams have been used for the same purpose, even though the syntax and semantics of activity diagrams has not been yet fully worked out. Nevertheless, activity diagrams seem very similar to Petri net semantics. UML, being the industry standard as a common object oriented modeling language needs a well-defined semantic base for its notation. Formalization of the graphical notation enables automated processing and analysis tasks. Petri nets can provide a formal semantic framework for the UML notations plus the behavioral modeling/analysis strength needed to system designers. This paper describes the methodology for creating the model of the RT application that would allow testing the correctness of the algorithm and the fulfillment of the time constraints at the design stage using UML and Petri Nets.
CWW elements to enrich SWOT analysis
Journal of Intelligent & Fuzzy Systems

Applied Sciences
Regardless of age, it is always important to detect deviations in long-term blood pressure from n... more Regardless of age, it is always important to detect deviations in long-term blood pressure from normal levels. Continuous monitoring of blood pressure throughout the day is even more important for elderly people with cardiovascular diseases or a high risk of stroke. The traditional cuff-based method for blood pressure measurements is not suitable for continuous real-time applications and is very uncomfortable. To address this problem, continuous blood pressure measurement methods based on photoplethysmogram (PPG) have been developed. However, these methods use specialized high-performance hardware and sensors, which are not available for common users. This paper proposes the continuous systolic blood pressure (SBP) estimation method based on PPG pulse wave steepness for low processing power wearable devices and evaluates its suitability using the commercially available CMS50FW Pulse Oximeter. The SBP estimation is done based on the PPG pulse wave steepness (rising edge angle) becaus...
Elektronika ir Elektrotechnika
Increasing resident's comfort and reducing energy costs have always been two primary objectives o... more Increasing resident's comfort and reducing energy costs have always been two primary objectives of intelligent lighting control systems. It is quite difficult to provide control satisfying the level of individual comfort, sufficient illumination and the energy reduction goals simultaneously. However, finding the balance between resident's preferred and recommended illumination for the current resident's activity may be beneficial. This paper addresses the problem of ensuring semi-autonomous assistance in controlling the intensity of light sources. The proposed decision making algorithm allows to provide gradual adaptation to the recommended illumination according the resident's activity. Resident's activity recognition is performed using one of the most popular models of deep learning, such as Convolutional Neural Networks (CNNs).

Applied Mathematical Modelling, 2017
The study presents an attempt to develop a model for the management of indoor air quality based o... more The study presents an attempt to develop a model for the management of indoor air quality based on real-time sensing of particulate matter (nano and micro particles) and volatile organic compounds. The development of the model used a grey box approach where the initial data on pollutant variation was collected during the experimental phase, and further applied to the pollutant mass balance model. The pollution sources have been analyzed in a controlled environment to obtain patterns of temporal variation, which have been approximated by mathematical functions. Approximations allowed the employment of pollutant mass balance model for determining the variation of pollutant source and further to modeling variation of pollutant concentration with changing air change rate. The proposed management approach can be applied to control indoor air quality in homes, assuring optimal utilization of the air handling unit in order to achieve the acceptable indoor air quality in the lowest time span and optimal energy use.
Uploads
Papers by Egidijus Kazanavicius