Aleksander Palkowski

A System for Heart Sounds Classification

The future of quick and efficient disease diagnosis lays in the development of reliable non-invasive methods. As for the cardiac diseases – one of the major causes of death around the globe – a concept of an electronic stethoscope equipped with an automatic heart tone identification system appears to be the best solution. Thanks to the advancement in technology, the quality of phonocardiography signals is no longer an issue. However, appropriate algorithms for auto-diagnosis systems of heart diseases that could be capable of distinguishing most of known pathological states have not been yet developed. The main issue is non-stationary character of phonocardiography signals as well as a wide range of distinguishable pathological heart sounds. In this paper a new heart sound classification technique, which might find use in medical diagnostic systems, is presented. It is shown that by combining Linear Predictive Coding coefficients, used for future extraction, with a classifier built upon combining Support Vector Machine and Modified Cuckoo Search algorithm, an improvement in performance of the diagnostic system, in terms of accuracy, complexity and range of distinguishable heart sounds, can be made. The developed system achieved accuracy above 93% for all considered cases including simultaneous identification of twelve different heart sound classes. The respective system is compared with four different major classification methods, proving its reliability.

Body surface area formulae: an alarming ambiguity

Body surface area (BSA) plays a key role in several medical fields, including cancer chemotherapy, transplantology, burn treatment and toxicology. BSA is often a major factor in the determination of the course of treatment and drug dosage. A series of formulae to simplify the process have been developed. Because easy-to-identify, yet general, body coefficient results of those formulae vary considerably, the question arises as to whether the choice of a particular formula is valid and safe for patients. Here we show that discrepancies between most of the known BSA formulae can reach 0.5 m2 for the standard adult physique. Although many previous studies have demonstrated that certain BSA formulae provide an almost exact fit with the patients examined, all of these studies have been performed on a limited and isolated group of people. Our analysis presents a broader perspective, considering 25 BSA formulae. The analysis revealed that the choice of a particular formula is a difficult task. Differences among calculations made by the formulae are so great that, in certain cases, they may considerably affect patients’ mortality, especially for people with an abnormal physique or for children.

Basic Hand Gestures Classification Based on Surface Electromyography

This paper presents an innovative classification system for hand gestures using 2-channel surface electromyography analysis. The system developed uses the Support Vector Machine classifier, for which the kernel function and parameter optimisation are conducted additionally by the Cuckoo Search swarm algorithm. The system developed is compared with standard Support Vector Machine classifiers with various kernel functions. The average classification rate of 98.12% has been achieved for the proposed method.

Using River Formation Dynamics Algorithm in Mobile Robot Navigation

River Formation Dynamics is a heuristic optimization algorithm based on the manner, in which drops of water form the river bed. The idea is to imitate the movement of drops on the edges between given nodes thus performing a search based on their height, which is modified through the mechanism of soil erosion and sediment deposition. In this way decreasing gradients are constructed, and these are followed by subsequent drops to compose new gradients and reinforce the best ones. This paper presents implementation of the method in wheeled mobile robot path searching task in an environment with obstacles. At first, a concise introduction to the RFD algorithm is presented, along with the problems requirements. In the main part a simulation test has been performed using the algorithm to find the shortest route from the starting point to the goal. The tests were made with different landscape layouts, measuring the computation time. The last section presents a real life implementation of the algorithm using a Pioneer P3-DX mobile robot. The validity of this solution is discussed, as well as any modifications, along with a brief comparison with a similar, popular search path algorithm.

Generating optimal paths in dynamic environments using River Formation Dynamics algorithm

Abstract The paper presents a comparison of four optimisation algorithms implemented for the purpose of finding the shortest path in static and dynamic environments with obstacles. Two classical graph algorithms – the Dijkstra complete algorithm and A* heuristic algorithm – were compared with metaheuristic River Formation Dynamics swarm algorithm and its newly introduced modified version. Moreover, another swarm algorithm has been compared – the Ant Colony Optimization and its modification. Terms and conditions of the simulation are thoroughly explained, paying special attention to the new, modified River Formation Dynamics algorithm. The algorithms were used for the purpose of generating the shortest path in three different types of environments, each served as a static environment and as a dynamic environment with changing goal or changing obstacles. The results show that the proposed modified River Formation Dynamics algorithm is efficient in finding the shortest path, especially when compared to its original version. In cases where the path should be adjusted to changes in the environment, calculations carried out by the proposed algorithm are faster than the A*, Dijkstra, and Ant Colony Optimization algorithms. This advantage is even more evident the more complex and extensive the environment is.

Application of 3D Whole Body Scanning in Research on Human Body Surface Area

Human body surface area (BSA) is one of the major parameters used in several medical fields. Its heterogeneity caused by individual human characteristics sustains a many-decades-long research on the matter. Today’s technology allows to create exact body models in mere seconds. However, an extensive research that includes scanning people with major disfigurements or people suffering from diseases, which often prevent their mobility, requires a specific approach. Here, we present the entire scanning procedure and graphical processing methods used in investigation of changes in BSA in function of different anthropometric parameters. The Artec 3D Eva hand-held scanner has been used as the measurement device. We performed a total of five scans for every subject—four for each limb and one for the main body part. After a series of processing methods, the resulting body model can be further used as an accurate basis for BSA formulae investigation.

Evaluation of Body Surface Area Formulae Based on 3D Body Scans

Human body surface area (BSA) is an established parameter for the calculation of chemotherapy drugs dosage, treatment of chronic hepatitis B, treatment of burns or for establishing a dosing regimen for antimicrobials. Although being a critical parameter, usage of modern 3D scanners, which would measure the exact BSA value, is often impossible in time-sensitive operations or for patients unable to withstand the scanning process. Therefore, over the last decades considerable research efforts have been devoted to development of simple formulae for BSA approximation. The formulae use a small number of state variables (weight, height, age, sex), which are intended to be easily obtained for every patient. The formulae parameters were estimated independently for isolated groups of subjects, which should rise suspicion whether any of the formulae used is indeed effective and safe for medical treatment. Here, we provide an extended analysis of 43 BSA formulae based on 152 patients scanned with a hand-held 3D scanner. Upon comparison of the real BSA values with estimations made by the formulae, we can conclude that most of the formulae exhibit a high relative BSA error, ranging from 9.83% to 43.27%.

A conception of pairwise comparisons model for selection of appropriate body surface area calculation formula

Body surface area (BSA) may be computed using a variety of formulas, but the computed BSA differs from real BSA values for particular subjects. This is presented in the paper by computing BSA values for selected subject and comparing them to the real BSA value obtained with the use of a 3D body scanner. The results show inequalities in the relevant BSA computing formulas. Hence, there is a need to determine a method that will allow to select the best formula for calculating BSA in a particular case. For this purpose, the pairwise comparisons (PC) method is suggested. This article presents a proposition of using consistency-driven PC, as well as the basic and most important aspects of using PC to determine the appropriate BSA calculation formula.