Wyniki wyszukiwania - BazTech

1

Wykorzystanie urządzeń mobilnych do akwizycji i przetwarzania sygnałów biomedycznych w czasie rzeczywistym

Harasimczuk Adrian, Sieczkowski Krzysztof

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2025

|

nr 1

21--27

PL

W artykule opisano projekt aplikacji mobilnej, dedykowanej dla urządzenia typu smartphone pracującego pod kontrolą systemu operacyjnego Android, przeznaczonej do akwizycji sygnałów biomedycznych. Wybranym sygnałem biomedycznym stosowanym w niniejszej pracy był sygnał PPG (ang. Photoplethysmogram), pozyskany z dedykowanego układu. Dane zbierane są przy pomocy modułu ESP 32 z podłączonym sensorem typu MA X30102. Następnie przesyłane są w czasie rzeczywistym do urządzenia mobilnego za pośrednictwem sieci Bluetooth Low Energy. Do przetwarzania danych biomedycznych wykorzystano procesy filtrowania sygnału, detekcji wartości szczytowych, wyliczania pulsu i saturacji. Użytkownik ma możliwość obserwacji wyników w czasie rzeczywistym na dedykowanych wykresach. Wykonane badania testowe potwierdzają poprawność działania aplikacji oraz zawierają porównanie efektów pracy z pomiarami wykonanymi za pomocą urządzeń komercyjnych typu pulsoksymetr i smartwatch.

EN

This paper describes the design of a mobile application for a smartphone device running the Android operating system with the intent of acquiring biomedical signals. The selected biomedical signal used in this work was the Photoplethysmogram (PPG ) signal, acquired via a dedicated Analog Front-End (AFE) chip. The data is collected using an ESP 32 module with a MA X30102 type AFE sensor connected. It is then transmitted in real time to a mobile device via Bluetooth Low Energy. Signal filtering, peak detection, as well as pulse and saturation calculation processes were used to process the biomedical data. The user can observe the results in real time on dedicated graphs. Conducted test studies confirm that the application functions correctly and include a comparison of the results with measurements made with commercial devices such as pulse oximeter and smartwatch.

2

Research on target positioning and signal processing based on the four-quadrant detector

Shen Wen, Fang Shaoxi, Gao Xu, Cui Xuejian, Feng Shuanglong

Opto-Electronics Review

|

2025

|

Vol. 33, No. 2

art. no. e154200

EN

In a target location system, the spot size and spot position of the laser light, which are received by a four-quadrant detector, are closely related to the accuracy of the target location. To acquire a pA-level current, according to the characteristics of the signals generated by the four-quadrant detector and the related signal processing circuits, this paper designs signal processing circuits with amplification multiples of 140 dB and 160 dB. Meanwhile, a T‑shaped compensation network is added to the circuit to solve the problem of bandwidth and gain not being increased simultaneously. Theoretical calculations and numerical simulations are carried out to verify this circuit gain, phase margin, and other parameters. Simulation results show that the bandwidth of the signal processing circuit with an amplification of 140 dB is 199.256 kHz, the bandwidth of the signal processing circuit with an amplification of 160 dB is 100 kHz, and the test bandwidth is 96 kHz in 140 dB, which provides a strong support for the calculation of the spot position of the four-quadrant detector.

3

Improve the safety of air transport, especially in militarized terrain, by use of side looking airborne radar and space time adaptive processing

Ślesicki Błażej, Ślesicka Anna, Kawalec Adam

Zeszyty Naukowe. Transport / Politechnika Śląska

|

2024

|

z. 123

335--346

EN

The paper explores the potential to enhance aviation safety, particularly in militarized regions, by outfitting aircraft with Side Looking Airborne Radar (SLAR) and employing space-time adaptive processing (STAP) algorithms. The research objective revolves around implementing a model of side-looking airborne radar and the corresponding STAP algorithms. This technology enables the detection of slow-moving targets amidst strong interference, encompassing both passive (clutter) and active (jammer) elements. Slow-moving targets relative to the aircraft's speed include tanks, combat vehicles, command vehicles, artillery, and logistical assets of enemy forces. The theoretical framework of space-time adaptive processing is presented, elucidating the sequential steps of the classical Sample Matrix Inversion Space-Time Adaptive Processing (SMI STAP) algorithm. The paper underscores the significance of characteristic parameters delineating a linear STAP processor. The proposed solution facilitates the detection of enemy combat measures and enhances aviation safety. It outlines a radar model installed beneath the aircraft's fuselage and elucidates algorithms for space-time adaptive processing of radar signals. The simulations conducted within the article were executed using the MATLAB environment. The simulation results indeed suggest that the proposed solution holds promise for deployment in equipping aircraft of one's own military and those engaged in operations within conflict zones. This paper stands as one of the few contributions in the literature addressing the augmentation of aircraft safety through radar and space-time adaptive processing.

4

A new result on description of the spectrum of a sampled signal

Borys A.

TransNav : International Journal on Marine Navigation and Safety of Sea Transportation

|

2024

|

Vol. 18 No. 2

393--397

EN

It is explained, in introduction of this paper, why the description of the output signal at an A/D converter in the form that is presented in such respected textbooks as: a one written by Prandoni and Vetterli, and another one by van de Plassche is appropriate and correct. Unlike all others, especially those using in it the so-called comb of Dirac deltas. The latter ones do not lead to getting a correct formula for the spectrum of the output waveform of an A/D converter, or they yield no formula at all. Using the description of the A/D output signal in form of a step function (as in the textbooks mentioned above), a new, correct formula for calculating the spectrum of the sampled signal is derived in this paper. It is a revised version of the formula currently used in the literature, that is of the so-called Discrete-Time Fourier Transform (DTFT), and it is a product of this DTFT and a certain correction factor. Finally, some literature items are referred to in which the designers of integrated circuits (containing A/D converters) point out discrepancies that arise in designs when the multiplying factor mentioned above is not taken into account.

5

Porównanie jakości audio wybranej metody znakowania wodnego dla sygnału mowy, muzyki klasycznej i popularnej

Jekateryńczuk Gabriel, Piotrowski Zbigniew

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2024

|

nr 4

333--336

PL

W artykule przedstawiono szczegółową analizę oraz porównanie metody znakowania wodnego sygnałów audio za pomocą metody cyfrowego przetwarzania sygnałów bez przetwarzania sygnałów sieciami neuronowymi. Skupiono się na zastosowaniu dyskretnej transformaty falkowej – DWT oraz dyskretnej transformaty kosinusowej - DCT. Głównym celem badań jest ocena i porównanie miar jakości znaku wodnego osadzonego w różnych rodzajach treści audio, takich jak sygnał mowy, muzyka popularna i muzyka klasyczna.

EN

The article presents a detailed analysis and comparison of watermarking method for audio signals using digital signal processing without signal processing by neural networks. The focus is on the application of discrete wavelet transform - DWT and discrete cosine transform - DCT. The main goal of the research is to evaluate and compare the quality measures of the watermark embedded in various types of audio content, such as speech signal, popular music, and classical music.

6

Deep learning in the classification and recognition of cardiac activity patterns

Jeleń Łukasz, Ciskowski Piotr, Kluwak Konrad

International Journal of Electronics and Telecommunications

|

2024

|

Vol. 70, No. 1

79--85

EN

Electrocardiography is an examination performed frequently in patients experiencing symptoms of heart disease. Upon a detailed analysis, it has shown potential to detect and identify various activities. In this article, we present a deep learning approach that can be used to analyze ECG signals. Our research shows promising results in recognizing activity and disease patterns with nearly 90% accuracy. In this paper, we present the early results of our analysis, indicating the potential of using deep learning algorithms in the analysis of both onedimensional and two–dimensional data. The methodology we present can be utilized for ECG data classification and can be extended to wearable devices. Conclusions of our study pave the way for exploring live data analysis through wearable devices in order to not only predict specific cardiac conditions, but also a possibility of using them in alternative and augmented communication frameworks.

7

Wavelet analysis of ultrasonic lamb wave displacements in three-layer adhesive plates: continuous wavelet transform (CWT) versus the semi-analytical finite element method (SAFEM)

Yacoubi Abdelali, Jabiri Ayoub, Azkour Mustapha, Mandry Rachid, Allami Mhammed El

Engineering Transactions

|

2024

|

Vol. 72, iss. 2

203--224

EN

In this paper, we aim to identify the most appropriate mother wavelet for analyzing the displacements of ultrasonic guided waves in tri-layered adhesive plates.We determine the group velocities of a given mode using various mother wavelets. The precision of each mother wavelet is evaluated by comparing the values of the group velocities with those found by the semianalytical finite element method (SAFEM). The most appropriate mother wavelet function can then be used to study tri-layered adhesive plates with defects.

8

Vibration Signal Processing for Multirotor UAVs Fault Diagnosis: Filtering or Multiresolution Analysis

Al-Haddad Luttfi A., Giernacki Wojciech, Shandookh Ahmed A., Jaber Alaa Abdulhady, Puchalski Radosław

Eksploatacja i Niezawodność

|

2024

|

Vol. 26, no. 1

art. no. 176318

EN

In the modern technological advancements, Unmanned Aerial Vehicles (UAVs) have emerged across diverse applications. As UAVs evolve, fault diagnosis witnessed great advancements, with signal processing methodologies taking center stage. This paper presents an assessment of vibration-based signal processing techniques, focusing on Kalman filtering (KF) and Discrete Wavelet Transform (DWT) multiresolution analysis. Experimental evaluation of healthy and faulty states in a quadcopter, using an accelerometer, are presented. The determination of the 1024 Hz sampling frequency is facilitated through finite element analysis of 20 mode shapes. KF exhibits commendable performance, successfully segregating faulty and healthy peaks within an acceptable range. While thesix-level multi-decomposition unveils good explanations for fluctuations eluding KF. Ultimately, both KF and DWT showcase high-performance capabilities in fault diagnosis. However, DWT shows superior assessment precision, uncovering intricate details and facilitating a holistic understanding of fault-related characteristics.

9

Probabilistic and explainable modeling of Phase-Phase Cross-Frequency Coupling patterns in EEG. Application to dyslexia diagnosis

Castillo-Barnes Diego, Gallego-Molina Nicolás J., Formoso Marco A., Ortiz Andrés, Figueiredo Patrícia, Luque Juan L.

Biocybernetics and Biomedical Engineering

|

2024

|

Vol. 44, no. 4

814--823

EN

This work explores the intricate neural dynamics associated with dyslexia through the lens of Cross-Frequency Coupling (CFC) analysis applied to electroencephalography (EEG) signals evaluated from 48 seven-year-old Spanish readers from the LEEDUCA research platform. The analysis focuses on CFS (Cross-Frequency phase Synchronization) maps, capturing the interaction between different frequency bands during low-level auditory processing stimuli. Then, making use of Gaussian Mixture Models (GMMs), CFS activations are quantified and classified, offering a compressed representation of EEG activation maps. The study unveils promising results specially at the Theta-Gamma coupling (Area Under the Curve = 0.821), demonstrating the method’s sensitivity to dyslexia-related neural patterns and highlighting potential applications in the early identification of dyslexic individuals.

10

Drone Flight Detection at an Entrance to a Beehive Based on Audio Signals

Libal Urszula, Biernacki Paweł

Archives of Acoustics

|

2024

|

Vol. 49, nr 3

459--468

EN

Spotting a significant number of drones flying near the entrance of a beehive during late Spring could indicate the occurrence of swarming mood, as the the surge in drone presence is related to an overcrowded hive. Swarming refers to a natural reproductive process witnessed in honey bees, wherein half of the bee colony departs from their hive alongside the aging queen. In the paper, we propose an early swarming detection mechanism that relies on the behavior of the drones. The proposed method is based on audio signals registered in a close proximity to the beehive entrance. A comparative study was performed to find the most effective preprocessing method for the audio signals for the detection problem. We have compared the results for three different power spectrum density coefficients estimation methods, which are used as an input of an autoencoder neural network to discriminate drones from worker bees. Through simulations employing real-life signals, it has been demonstrated that drone detection based solely on audio signals is indeed feasible. The attained level of detection accuracy enables the creation of an efficient alarm system for beekeepers.

11

Filtering strategies for smartphone emitted digital signals

Obretin Alexandru Marius, Cornea Andreea-Alina

Applied Computer Science

|

2024

|

Vol. 20, no. 1

139--156

EN

In today's digitalized and technology-driven society, where the number of IoT devices and the volume of collected data is exponentially increasing, the use of sensor data has become a necessity in certain fields of activity. This paper presents a concise history of sensor evolution and specialization, with a focus on the sensors used for localization, particularly those present in microelectromechanical systems (MEMS) that make up inertial measurement units. The study starts with a general overview and progresses towards a more specific analysis of data sets collected from an accelerometer. In the materials and methods section, it emphasizes the imperfect nature of sensor measurements and the need to filter digital signals. Three different digital signal filtering techniques belonging to distinct filter categories are comparatively analyzed, with each technique having its own particularities, advantages and disadvantages. The analysis considers the effectiveness in reducing measurement errors, the impact of the filtering process on the original signal, the ability to highlight the underlying phenomenon, as well as the performance of the analyzed algorithms. The ultimate purpose of this article is to determine which filtration method is best suited for the signal at hand in the context of an indoor localization application.

12

Investigating Vibration Transmission in Cargo Containers During Heavy-Duty Off-Road Transport

Vilau Radu, Voicu Daniela, Stoica Ramona-Monica, Ciobotaru Ticusor, Rykała Magdalena, Rykała Łukasz

Advances in Science and Technology. Research Journal

|

2024

|

Vol. 18, no. 4

312--328

EN

This article focuses on the investigation and analysis of vibrations transmitted to cargo during off-road transportation, with particular emphasis on the impact of vehicle and road surface interactions. The primary objective of the research is to quantify and characterize the amplitudes of vibrations generated by the interaction between heavy-duty truck tires and rough terrain, and their subsequent transmission to cargo containers. To achieve this, a virtual model of a tactical transportation truck was created using TruckSim software. Two characteristic off-road tracks were simulated, based on driving conditions typically experienced by heavy-duty vehicles in demanding logistical scenarios. The experimental validation of the virtual model was conducted using a heavy-duty truck outfitted with a 20 ft (6096 mm) cargo container. The results of the work include recorded acceleration data, suspension behavior, and the maximum driving speed at which the vehicle remained stable on both tracks. Moreover, the work is a direct response to the needs of the automotive industry and the military.

13

Analiza porównawcza metod estymacji macierzy kowariancji zakłóceń w przestrzenno-czasowym adaptacyjnym przetwarzaniu sygnałów

Ślesicki Błażej, Ślesicka Anna, Kawalec Adam, Żmigrodzka Małgorzata

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2023

|

nr 4

297--300

PL

Artykuł zawiera analizę porównawczą statystycznych i niestatystycznych metod estymacji macierz kowariancji zakłóceń w przestrzenno-czasowym adaptacyjnym przetwarzaniu (ang. Space-Time Adaptive Processing STAP) sygnału radarowego dla modelu radaru MIMO (ang. Multiple Input Multiple Output). Zaprezentowano istotę, przebieg algorytmu STAP oraz właściwości najnowszych typów metod estymacji macierzy kowariancji zakłóceń. Autorzy przytoczyli w artykule również swój wkład w rozwój tej technologii.

EN

The article presents a comparative analysis of statistical and non-statistical methods of estimating the clutter covariance matrix in Space-Time Adaptive Processing by using Multiple Input Multiple Output radar geometry. In addition, the properties and stages of the STAP algorithm are presented, as well as the main features of the latest methods for estimating the clutter covariance matrix. Moreover, the authors referenced their contributions to the development of this technology in the article.

14

Wykrywanie zdarzeń awaryjnych w silnikach indukcyjnych na podstawie analizy danych z rejestratorów zakłóceń

Sajewicz Dariusz, Łaguna Wojciech, Chmielak Waldemar

Przegląd Elektrotechniczny

|

2023

|

R. 99, nr 1

207--210

PL

Celem artykułu jest wykazanie skuteczności nowo opracowanych metod detekcji uszkodzeń opartych na analizie danych z rejestratorów zakłóceń. W trakcie prac badawczych wyekstrahowano najbardziej istotne cechy sygnałów prądów w dziedzinie częstotliwości. Pozyskane cechy stanowiły podstawę budowy probabilistycznego klasyfikatora zdarzeń awaryjnych. Detekcja uszkodzeń dotyczy wykrywania: pękniętych prętów wirnika i stopnia jego degradacji oraz awarii łożysk na wale silnika. Przeprowadzone badania potwierdzają wysoką skuteczność wykrywania uszkodzeń we wszystkich rozpatrywanych obszarach.

EN

The aim of this paper is to demonstrate the effectiveness of developed fault detection methods based on the analysis of data from fault recorders. During the research work, the most significant features of current signals in the frequency domain were extracted. The extracted features provided the base for building a probabilistic classifier of fault incidents. The fault detection concerned the detection of cracked rotor cages and the degree of its degradation as well as the failure of bearings on the motor shaft. The conducted research confirms the high efficiency of detection faults in all areas concerned.

15

Correlation between linear and non-linear vibroartrographic parameters

Kręcisz Krzysztof, Bączkowicz Dawid, Łysiak Adam, Szmajda Mirosław, Kawala-Sterniuk Aleksandra

Przegląd Elektrotechniczny

|

2023

|

R. 99, nr 7

180--185

EN

In this paper the authors compared linear and non-linear features of the vibroarthrogram (VAG) in knee joint context. There were 220 healthy participants in the study, divided into five age-related groups (third, fourth, fifth, sixth and seventh decade of life). Four linear features (i.e. variance of the mean squares, amplitude, and two frequency parameters) and two non-linear features (i.e. multi-scale entropy and recurrence rate) were correlated with each other. Correlations within linear and non-linear groups proved to be positive, while inter-group correlations turn out to be negative. Also, in the context of age differentiation, recurrence rate feature proved to be the most informative one.

PL

W niniejszej pracy autorzy porównali liniowe i nieliniowe cechy wibroartrogramu (VAG) w kontekście stawu kolanowego. W badaniuwzięło udział 220 zdrowych uczestników, podzielonych na pięć grup wiekowych (trzecia, czwarta, piąta, szósta i siódma dekadażycia). Cztery cechy˙ liniowe (tj. wariancja średnich kwadratów, amplituda i dwa parametry częstotliwości) oraz dwie cechy nieliniowe (tj. wieloskalowa entropia i częstość nawrotów) były ze sobą skorelowane. Korelacje w obrębie grup liniowych i nieliniowych okazały się dodatnie, natomiast korelacje międzygrupowe okazały się ujemne. Również w kontekście zróżnicowania wieku cecha częstości nawrotów okazała się najbardziej informatywna.

16

Non-invasive method for blood glucose monitoring using ECG signal

Fellah Arbi Khadidja, Soulimane Sofiane, Saffih Faycal

Polish Journal of Medical Physics and Engineering

|

2023

|

Vol. 29, Iss. 1

1--9

EN

Introduction: Tight glucose monitoring is crucial for diabetic patients by using a Continuous Glucose Monitor (CGM). The existing CGMs measure the Blood Glucose Concentration (BGC) from the interstitial fluid. These technologies are quite expensive, and most of them are invasive. Previous studies have demonstrated that hypoglycemia and hyperglycemia episodes affect the electrophysiology of the heart. However, they did not determine a cohort relationship between BGC and ECG parameters. Material and method: In this work, we propose a new method for determining the BGC using surface ECG signals. Recurrent Convolutional Neural Networks (RCNN) were applied to segment the ECG signals. Then, the extracted features were employed to determine the BGC using two mathematical equations. This method has been tested on 04 patients over multiple days from the D1namo dataset, using surface ECG signals instead of intracardiac signal. Results: We were able to segment the ECG signals with an accuracy of 94% using the RCNN algorithm. According to the results, the proposed method was able to estimate the BGC with a Mean Absolute Error (MAE) of 0.0539, and a Mean Squared Error (MSE) of 0.1604. In addition, the linear relationship between BGC and ECG features has been confirmed in this paper. Conclusion: In this paper, we propose the potential use of ECG features to determine the BGC. Additionally, we confirmed the linear relationship between BGC and ECG features. That fact will open new perspectives for further research, namely physiological models. Furthermore, the findings point to the possible application of ECG wearable devices for non-invasive continuous blood glucose monitoring via machine learning.

17

Capon-like method for direction of arrival estimation using TDM MIMO radar

Ślesicka Anna, Kawalec Adam, Ślesicki Błażej

Metrology and Measurement Systems

|

2023

|

Vol. 30, nr 3

481--497

EN

This paper presents a novel measurement method and briefly discusses the basic properties of direction of arrival (DoA) measurement in a multiple-input multiple-output (MIMO) radar system by using orthogonality with time-division multiplexing (TDM), where only one transmitting antenna element is active in each time slot. This paper presents the mathematical model of the TDM-MIMO radar operating at 10 GHz, transmitting a string of pulses, the method of transmitting and receiving the signal, and the method of measuring the angle of arrival of the signal based on the use of the Capon algorithm and its modifications. Finally, the correctness of the theory, algorithm and method of measuring the direction of arrival of the signal is verified by experimental simulation. The work discussed in this paper is of great significance to practically demonstrate the capabilities of the TDM MIMO radar sensor in practical implementations like reconnaissance and electronic warfare systems.

18

Research on vibration-based early diagnostic system for excavator motor bearing using 1-D CNN

Yandagsuren Dorjsuren, Kurauchi Tatsuki, Toriya Hisatoshi, Ikeda Hajime, Adachi Tsuyoshi, Kawamura Youhei

Journal of Sustainable Mining

|

2023

|

Vol. 22, iss. 1

65--80

EN

In mining, super-large machines such as rope excavators are used to perform the main mining operations. A rope excavator is equipped with motors that drive mechanisms. Motors are easily damaged as a result of harsh mining conditions. Bearings are important parts in a motor; bearing failure accounts for approximately half of all motor failures. Failure reduces work efficiency and increases maintenance costs. In practice, reactive, preventive, and predictive maintenance are used to minimize failures. Predictive maintenance can prevent failures and is more effective than other maintenance. For effective predictive maintenance, a good diagnosis is required to accurately determine motor-bearing health. In this study, vibration-based diagnosis and a one-dimensional convolutional neural network (1-D CNN) were used to evaluate bearing deterioration levels. The system allows for early diagnosis of bearing failures. Normal and failure-bearing vibrations were measured. Spectral and wavelet analyses were performed to determine the normal and failure vibration features. The measured signals were used to generate new data to represent bearing deterioration in increments of 10%. A reliable diagnosis system was proposed. The proposed system could determine bearing health deterioration at eleven levels with considerable accuracy. Moreover, a new data mixing method was applied.

19

EEG signal analysis for monitoring concentration of operators

Rykała Łukasz

Journal of Automation Mobile Robotics and Intelligent Systems

|

2023

|

Vol. 17, No. 1

30--39

EN

Often, operators of machines, including unmanned ground vehicles (UGVs) or working machines, are forced to work in unfavorable conditions, such as high tem‐ peratures, continuously for a long period of time. This has a huge impact on their concentration, which usu‐ ally determines the success of many tasks entrusted to them. Electroencephalography (EEG) allows the study of the electrical activity of the brain. It allows the determination, for example, of whether the operator is able to focus on the realization of his tasks. The main goal of this article was to develop an algorithm for determining the state of brain activity by analyzing the EEG signal. For this purpose, methods of EEG sig‐ nal acquisition and processing were described, including EEG equipment and types and location of electrodes. Particular attention was paid to EEG signal acquisition, EEG signal artifacts, and disturbances, and elements of the adult’s correct EEG recording were described in detail. In order to develop the algorithm mentioned, basic types of brain waves were discussed, and exem‐ plary states of brain activity were recorded. The influ‐ ence of technical aspects on the recording of EEG sig‐ nals was also emphasized. Additionally, a block diagram was created which is the basis for the operation of the said algorithm. The LabVIEW environment was used to implement the created algorithm. The results of the research showing the operation of the developed EEG signal analyzer were also presented. Based on the results of the study, the EEG analyzer was able to accurately determine the condition of the examined person and could be used to study the concentration of machine operators.

20

Automatic detection of Alzheimer's disease based on artificial intelligence

Benba Achraf, Abdelilah Kerchaoui

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2023

|

T. 13, nr 1

18--21

EN

Alzheimer's disease is a neurodegenerative disease that progressively destroys neurons through the formation of platelets that prevent communication between neurons. The study carried out in this project aims to find a precise and relevant diagnostic solution based on artificial intelligence and which helps in the early detection of Alzheimer's disease in order to stop its progression. The study went through a process of processing MRI images followed by training of three deep learning algorithms (VGG-19, Xception and DenseNet121) and finally by a step of testing and predicting the results. The results of the accuracy metric obtained for the three algorithms were respectively 98%, 95%, 91%.

PL

Choroba Alzheimera jest chorobą neurodegeneracyjną, która stopniowo niszczy neurony poprzez tworzenie płytek krwi, które uniemożliwiają komunikację między neuronami. Badania prowadzone w ramach tego projektu mają na celu znalezienie precyzyjnego i trafnego rozwiązania diagnostycznego opartego na sztucznej inteligencji, które pomoże we wczesnym wykryciu choroby Alzheimera w celu zatrzymania jej postępu. Badanie przeszło przez proces przetwarzania obrazów MRI, po którym następowało szkolenie trzech algorytmów głębokiego uczenia (VGG-19, Xception i DenseNet121), a na koniec etap testowania i przewidywania wyników. Wyniki metryki dokładności otrzymane dla trzech algorytmów wyniosły odpowiednio 98%, 95%, 91%.