Wyniki wyszukiwania - BazTech

1

Applicability of artificial intelligence in smart healthcare systems for automatic detection of Parkinson’s Disease

Pallathadka Harikumar, Padminivalli V. S.J.R.K., Vasavi M., Nancy P., Naved Mohd, Kumar Harish, Ray Samrat

Computer Assisted Methods in Engineering and Science

|

2024

|

Vol. 31, no. 2

175--185

EN

Parkinson’s disease is associated with memory loss, anxiety, and depression in the brain. Problems such as poor balance and difficulty during walking can be observed in addition to symptoms of impaired posture and rigidity. The field dedicated to making computers capable of learning autonomously, without having to be explicitly programmed, is known as machine learning. An approach to the diagnosis of Parkinson’s disease, which is based on artificial intelligence, is discussed in this article. The input for this system is provided through photographic examples of Parkinson’s disease patient handwriting. Received photos are preprocessed using the relief feature option to begin the process. This is helpful in the process of selecting characteristics for the identification of Parkinson’s disease. After that, the linear discriminant analysis (LDA) algorithm is employed to reduce the dimensions, bringing down the total number of dimensions that are present in the input data. The photos are then classified via radial basis function-support vector machine (SVM-RBF), k-nearest neighbors (KNN), and naive Bayes algorithms, respectively.

2

Comprehensive machine learning and deep learning approaches for Parkinson's disease classification and severity assessment

Majdoubi Oumaima, Benba Achraf, Hammouch Ahmed

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2023

|

T. 13, nr 4

15--20

EN

In this study, we aimed to adopt a comprehensive approach to categorize and assess the severity of Parkinson's disease by leveraging techniques from both machine learning and deep learning. We thoroughly evaluated the effectiveness of various models, including XGBoost, Random Forest, Multi-Layer Perceptron (MLP), and Recurrent Neural Network (RNN), utilizing classification metrics. We generated detailed reports to facilitate a comprehensive comparative analysis of these models. Notably, XGBoost demonstrated the highest precision at 97.4%. Additionally, we took a step further by developing a Gated Recurrent Unit (GRU) model with the purpose of combining predictions from alternative models. We assessed its ability to predict the severity of the ailment. To quantify the precision levels of the models in disease classification, we calculated severity percentages. Furthermore, we created a Receiver Operating Characteristic (ROC) curve for the GRU model, simplifying the evaluation of its capability to distinguish among various severity levels. This comprehensive approach contributes to a more accurate and detailed understanding of Parkinson's disease severity assessment.

PL

W tym badaniu naszym celem było przyjęcie kompleksowego podejścia do kategoryzacji i oceny ciężkości choroby Parkinsona poprzez wykorzystanie technik zarówno uczenia maszynowego, jak i głębokiego uczenia. Dokładnie oceniliśmy skuteczność różnych modeli, w tym XGBoost, Random Forest, Multi-Layer Perceptron (MLP) i Recurrent Neural Network (RNN), wykorzystując wskaźniki klasyfikacji. Wygenerowaliśmy szczegółowe raporty, aby ułatwić kompleksową analizę porównawczą tych modeli. Warto zauważyć, że XGBoost wykazał najwyższą precyzję na poziomie 97,4%. Ponadto poszliśmy o krok dalej, opracowując model Gated Recurrent Unit (GRU) w celu połączenia przewidywań z alternatywnych modeli. Oceniliśmy jego zdolność do przewidywania nasilenia dolegliwości. Aby określić ilościowo poziomy dokładności modeli w klasyfikacji chorób, obliczyliśmy wartości procentowe nasilenia. Ponadto stworzyliśmy krzywą charakterystyki operacyjnej odbiornika (ROC) dla modelu GRU, upraszczając ocenę jego zdolności do rozróżniania różnych poziomów nasilenia. To kompleksowe podejście przyczynia się do dokładniejszego i bardziej szczegółowego zrozumienia oceny ciężkości choroby Parkinsona.

3

Classification of Parkinson’s disease and other neurological disorders using voice features extraction and reduction techniques

Majdoubi Oumaima, Benba Achraf, Hammouch Ahmed

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2023

|

T. 13, nr 3

16--22

EN

This study aimed to differentiate individuals with Parkinson's disease (PD) from those with other neurological disorders (ND) by analyzing voice samples, considering the association between voice disorders and PD. Voice samples were collected from 76 participants using different recording devices and conditions, with participants instructed to sustain the vowel /a/ comfortably. PRAAT software was employed to extract features including autocorrelation (AC), cross-correlation (CC), and Mel frequency cepstral coefficients (MFCC) from the voice samples. Principal component analysis (PCA) was utilized to reduce the dimensionality of the features. Classification Tree (CT), Logistic Regression, Naive Bayes (NB), Support Vector Machines (SVM), and Ensemble methods were employed as supervised machine learning techniques for classification. Each method provided distinct strengths and characteristics, facilitating a comprehensive evaluation of their effectiveness in distinguishing PD patients from individuals with other neurological disorders. The Naive Bayes kernel, using seven PCA-derived components, achieved the highest accuracy rate of 86.84% among the tested classification methods. It is worth noting that classifier performance may vary based on the dataset and specific characteristics of the voice samples. In conclusion, this study demonstrated the potential of voice analysis as a diagnostic tool for distinguishing PD patients from individuals with other neurological disorders. By employing a variety of voice analysis techniques and utilizing different machine learning algorithms, including Classification Tree, Logistic Regression, Naive Bayes, Support Vector Machines, and Ensemble methods, a notable accuracy rate was attained. However, further research and validation using larger datasets are required to consolidate and generalize these findings for future clinical applications.

PL

Przedstawione badanie miało na celu różnicowanie osób z chorobą Parkinsona (PD) od osób z innymi zaburzeniami neurologicznymi poprzez analizę próbek głosowych, biorąc pod uwagę związek między zaburzeniami głosu a PD. Próbki głosowe zostały zebrane od 76 uczestników przy użyciu różnych urządzeń i warunków nagrywania, a uczestnicy byli instruowani, aby wydłużyć samogłoskę /a/ w wygodnym tempie. Oprogramowanie PRAAT zostało zastosowane do ekstrakcji cech, takich jak autokorelacja (AC), krzyżowa korelacja (CC) i współczynniki cepstralne Mel (MFCC) z próbek głosowych. Analiza składowych głównych (PCA) została wykorzystana w celu zmniejszenia wymiarowości cech. Jako techniki nadzorowanego uczenia maszynowego wykorzystano drzewa decyzyjne (CT), regresję logistyczną, naiwny klasyfikator Bayesa (NB), maszyny wektorów nośnych (SVM) oraz metody zespołowe. Każda z tych metod posiadała swoje unikalne mocne strony i charakterystyki, umożliwiając kompleksową ocenę ich skuteczności w rozróżnianiu pacjentów z PD od osób z innymi zaburzeniami neurologicznymi. Naiwny klasyfikator Bayesa, wykorzystujący siedem składowych PCA, osiągnął najwyższy wskaźnik dokładności na poziomie 86,84% wśród przetestowanych metod klasyfikacji. Należy jednak zauważyć, że wydajność klasyfikatora może się różnić w zależności od zbioru danych i konkretnych cech próbek głosowych. Podsumowując, to badanie wykazało potencjał analizy głosu jako narzędzia diagnostycznego do rozróżniania pacjentów z PD od osób z innymi zaburzeniami neurologicznymi. Poprzez zastosowanie różnych technik analizy głosu i wykorzystanie różnych algorytmów uczenia maszynowego, takich jak drzewa decyzyjne, regresja logistyczna, naiwny klasyfikator Bayesa, maszyny wektorów nośnych i metody zespołowe, osiągnięto znaczący poziom dokładności. Niemniej jednak, konieczne są dalsze badania i walidacja na większych zbiorach danych w celu skonsolidowania i uogólnienia tych wyników dla przyszłych zastosowań klinicznych.

4

A novel Parkinson's disease detection algorithm combined EMD, BFCC, and SVM classifier

Boualoulou Nouhaila, Mounia Miyara, Nsiri Benayad, Behoussine Drissi Taoufiq

Diagnostyka

|

2023

|

Vol. 24, No. 4

art. no. 2023404

EN

Identifying and assessing Parkinson's disease in its early stages is critical to effectively monitoring the disease's progression. Methodologies based on machine learning enhanced speech analysis are gaining popularity as the potential of this field is revealed. Acoustic features, in particular, are used in a variety of algorithms for machine learning and could serve as indicators of the general health of subjects' voices. In this research paper, a novel method is introduced for the automated detection of Parkinson's disease through speech signal analysis, a support vector machines classifier (SVM) and an Artificial Neural Network (ANN) are used to evaluate and classify the data based on two acoustic features: Bark Frequency Cepstral Coefficients (BFCC) and Mel Frequency Cepstral Coefficients (MFCC). These features are extracted from the denoised signals using Empirical Mode Decomposition (EMD). The most relevant results obtained for a dataset of 38 participants are by the BFCC coefficients with an accuracy up to 92.10%. These results confirm that EMD-BFCC-SVM method can contribute to the detection of Parkinson's disease.

5

Application of imaging techniques to objectify the Finger Tapping test used in the diagnosis of Parkinson's disease

Jakubowski Jacek, Potulska-Chromik Anna, Chmielińska Jolanta, Nojszewska Monika, Kostera-Pruszczyk Anna

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2023

|

Vol. 71, nr 2

art. no. e144886

EN

Finger tapping is one of the standard tests for Parkinson's disease diagnosis performed to assess the motor function of patients' upper limbs. In clinical practice, the assessment of the patient's ability to perform the test is carried out visually and largely depends on the experience of clinicians. This article presents the results of research devoted to the objectification of this test. The methodology was based on the proposed measurement method consisting in frame processing of the video stream recorded during the test to determine the time series representing the distance between the index finger and the thumb. Analysis of the resulting signals was carried out in order to determine the characteristic features that were then used in the process of distinguishing patients with Parkinson's disease from healthy cases using methods of machine learning. The research was conducted with the participation of 21 patients with Parkinson's disease and 21 healthy subjects. The results indicate that it is possible to obtain the sensitivity and specificity of the proposed method at the level of approx. 80 %. However, the patients were in the so-called ON phase when symptoms are reduced due to medication, which was a much greater challenge compared to analyzing signals with clearly visible symptoms as reported in related works.

6

Rola metali w rozwoju choroby Alzheimera i Parkinsona

Żygowska Justyna, Szymańska Aneta

Wiadomości Chemiczne

|

2022

|

[Z] 76, 1-2

1--25

EN

Neurodegenerative diseases are the consequence of progressive brain degeneration caused by the death of nerve cells. Many factors that influence the neurodegeneration development are still not fully known. A lot of studies indicate the contribution of metal ions in this process. Copper, zinc, and iron are trace elements essential for proper functioning of the body. They are part of many enzymes participating in the transmission of the nerve signals, electrons transport, neurotransmitters and nucleic acids synthesis, and oxygen storage. Disorder of metals homeostasis leads to the development of severe diseases and nervous system degenerations. An excess of copper and iron ions causes a significant increase in cellular oxidative stress. Metals catalyze the reactions of free radicals formation that destroy proteins, lipids, and nucleic acids. High concentration of copper and iron ions were found in the deposits of amyloidogenic proteins. Amyloid β (Alzheimer disease) and α synuclein (Parkinson disease) have ions binding chain structures. The metal-protein interaction increases oligomerization speed in vitro. A lot of evidence suggests that the disorder of Cu, Zn and Fe homeostasis accelerates the progress of brain neurodegeneration. Human organism contains many metals, which are not needed for the proper functioning of the body, e.g. aluminum. Al binds to nucleic acids causing an increase in cellular oxidative stress and initiating proteins oligomerization. The presence of aluminum is also considered to be disadvantageous for the nervous system. The lack of medicines for neurodegenerative diseases forces us to search for new therapies. The development of degenerations could be slowed down by chelators of toxic metals, but first, these diseases must be better understood. Adverse effects of high concentration of metal ions on brain functioning are not fully known. This knowledge is necessary to find effective drugs.

7

Parametric analysis of pilot voice signals in Parkinson’s disease diagnostics

Majda-Zdancewicz Ewelina, Potulska-Chromik Anna, Nojszewska Monika, Kostera-Pruszczyk Anna

Journal of Automation, Electronics and Electrical Engineering

|

2022

|

Vol. 4, No. 1

21--28

EN

Parkinson's disease (PD) is a neurodegenerative disease of the central nervous system (CNS) characterized by the progressive loss of dopaminergic neurons in the substantia nigra. The article describes an analysis of pilot voice signal analysis in Parkinson's disease diagnostics. Frequency domain signal analysis was mainly used to assess the state of a patient's voice apparatus in order to support PD diagnostics. The recordings covered uttering the “a” sound at least twice with extended phonation. The research utilized real recordings acquired in the Department of Neurology at the Medical University of Warsaw, Poland. Spectral speech signal coefficients may be determined based on different defined frequency scales. The authors used four frequency scales: linear, Mel, Bark and ERB . Spectral descriptors have been defined for each scales which are widely used in machine and deep learning applications, and perceptual analysis. The usefulness of extracted features was assessed taking into account various methods. The discriminatory ability of individual coefficients was evaluated using the Fisher coefficient and LDA technique.. The results of numerical experiments have shown different efficiencies of the proposed descriptors using different frequencies scales.

PL

Choroba Parkinsona (PD) jest neurodegeneracyjną chorobą ośrodkowego układu nerwowego charakteryzującą się postępującą utratą neuronów dopaminergicznych w istocie czarnej. W artykule opisano analizę rejestracji pilotażowych sygnałów głosu w diagnostyce choroby Parkinsona. Rejestracji podlegało co najmniej dwukrotnie wypowiadanie głoski "a” o przedłużonej fonacji. Do badań wykorzystano nagrania zarejestrowane w Katedrze i Klinice Neurologii Warszawskiego Uniwersytetu Medycznego w Warszawie. Do oceny stanu aparatu głosu pacjenta celem wsparcia diagnostyki choroby Parkinsona wykorzystano w głównej mierze analizę sygnału w dziedzinie częstotliwości. Autorzy zastosowali cztery skale częstości: liniową, skalę typu Mel, skalę typu Bark oraz skalę typu ERB. Dla każdej z tych skali zdefiniowali deskryptory spektralne szeroko stosowane w aplikacjach uczenia maszynowego i głębokiego uczenia się oraz w analizie percepcyjnej. Ocena przydatności wyekstrahowanych cech została zrealizowana z uwzględnieniem różnych metod. Wykorzystano metodą oceny jakości cech przy użyciu współczynnika istotności Fischera oraz analizę LDA. Wyniki eksperymentów numerycznych wykazały różne wydajności proponowanych deskryptorów przy użyciu różnych skal częstości.

8

Analysis of hand and face images for the purpose of engineering support for Parkinson's disease diagnosis

Białek Kamila, Potulska-Chromik Anna, Jakubowski Jacek, Nojszewska Monika, Kostera-Pruszczyk Anna

Journal of Automation, Electronics and Electrical Engineering

|

2022

|

Vol. 4, No. 1

13--20

EN

Engineering support in the field of recognizing Parkinson's disease against the background of other diseases, its progression and monitoring the effectiveness of drugs is currently widely implementedas part of work devoted to the use of recording and analysis devices equipped with sensors of movement parameters attached to the patient's body, e.g. accelerometers and gyroscopes. This material touches on an alternative approach, in which the concept of using techniques for processing selected image data obtained during a clinical examination evaluating a patient using the unified UPDRS number scale is proposed. The research was conducted on a material that corresponded to selected components of the scale and included images of faces recorded in the visible light range and images of the outer surfaces of the hand recorded with a thermal imaging camera.This was aimed at assessing the possibility of differentiating personsin terms of detecting Parkinson's disease on the basis of registered modalities. Thus, tasks aimed at developing characteristics important in the binary classification process were carried out. The assessment of features was made in a modality-dependent manner based on available tools in the field of statistics and machine learning.

PL

Wsparcie inżynierskie w zakresie rozpoznawania choroby Parkinsona na tle innych chorób, jej progresji oraz monitorowania skuteczności leków jest obecnie szeroko realizowane w ramach prac poświęconych wykorzystaniu urządzeń rejestrujących i analizujących wyposażonych w sensory parametrów ruchu przymocowanych do ciała pacjenta, np. akcelerometry i żyroskopy. W prezentowanej pracy przedstawiono alternatywne podejście, w którym proponuje się koncepcję wykorzystania technik przetwarzania wybranych danych obrazowych uzyskanych podczas badania klinicznego oceniającego pacjenta za pomocą ujednoliconej skali liczbowej UPDRS. Badania przeprowadzono na materiale, który odpowiadał wybranym składowym skali i obejmował obrazy twarzy utrwalone w zakresie światła widzialnego oraz obrazy zewnętrznych powierzchni dłoni rejestrowane kamerą termowizyjną. Wykonane badania miały na celu ocenę możliwości różnicowania osób pod względem wykrywania choroby Parkinsona na podstawie zarejestrowanych metod. W ten sposób zrealizowano zadania mające na celu opracowanie cech istotnych w procesie klasyfikacji binarnej. Ocena cech została dokonana w sposób zależny od modalności w oparciu o dostępne narzędzia z zakresu statystyki i uczenia maszynowego.

9

DaTSCAN – Radioizotopowa diagnostyka układu dopaminergicznego

Mackiewicz Patrycja, Krzemiński Jakub, Mączewska Joanna, Królicki Leszek

Inżynier i Fizyk Medyczny

|

2022

|

Vol. 11, nr 4

321--326

PL

Układ dopaminergiczny jest najczęściej badanym układem receptorowym z zastosowaniem metod medycyny nuklearnej. (...) Częstość występowania PD szacuje się na 20-30% [1]. W przebiegu choroby Parkinsona występuje zanik neuronów dopaminergicznych w obrębie istoty czarnej z formowaniem wewnątrzkomórkowo ciałek Lewy’ego. Uszkodzeniu ulegają również – w mniejszym stopniu – neurony serotoninergiczne jąder szwu, noradrenergiczne w obrębie jądra sinawego, a także komórki cholinergiczne jąder podstawy. Zanik komórek dopaminergicznych dotyczy głównie neuronów istoty czarnej położonych brzuszno-bocznie, tworzących drogę biegnącą do części grzbietowej skorupy. Neurony istoty czarnej położone grzbietowo- -przyśrodkowo, biegnące do głowy jądra ogoniastego, ulegają degeneracji dopiero w późnym okresie choroby. Mechanizm ten tłumaczy charakter zmian obserwowanych w badaniach obrazowych.

10

Enhanced decision tree induction using evolutionary techniques for Parkinson’s disease classification

Ghane Mostafa, Ang Mei Choo, Nilashi Mehrbakhsh, Sorooshian Shahryar

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 3

902--920

EN

The diagnosis of Parkinson’s disease (PD) is important in neurological pathology for appropriate medical therapy. Algorithms based on decision tree induction (DTI) have been widely used for diagnosing PD through biomedical voice disorders. However, DTI for PD diagnosis is based on a greedy search algorithm which causes overfitting and inferior solutions. This paper improved the performance of DTI using evolutionary-based genetic algorithms. The goal was to combine evolutionary techniques, namely, a genetic algorithm (GA) and genetic programming (GP), with a decision tree algorithm (J48) to improve the classification performance. The developed model was applied to a real biomedical dataset for the diagnosis of PD. The results showed that the accuracy of the J48, was improved from 80.51% to 89.23% and to 90.76% using the GA and GP, respectively.

11

End-to-end deep learning approach for Parkinson’s disease detection from speech signals

Quan Changqin, Ren Kang, Luo Zhiwei, Chen Zhonglue, Ling Yun

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 2

556--574

EN

More than 90% of patients with Parkinson’s disease suffer from hypokinetic dysarthria. This paper proposes a novel end-to-end deep learning model for Parkinson’s disease detection from speech signals. The proposed model extracts time series dynamic features using time-distributed two-dimensional convolutional neural networks (2D-CNNs), and then captures the dependencies between these time series using a one-dimensional CNN (1D-CNN). The performance of the proposed model was verified on two databases. On Database-1, the proposed model outperformed expert features-based machine learning models and achieved promising results, showing accuracies of 81.6% on the speech task of sustained vowel /a/ and 75.3% on the speech task of reading a short sentence (/si shi si zhi shi shi zi/) in Chinese. On Database-2, the proposed model was assessed on multiple sound types, including vowels, words, and sentences. An accuracy of up to 92% was obtained on the speech tasks, which included reading simple (/loslibros/) and complex (/viste/) sentences in Spanish. By visualizing the features generated by the model, it was found that the learned time series dynamic features are able to capture the characteristics of the reduced overall frequency range and reduced variability of Parkinson’s disease sounds, which are important clinical evidence for detecting Parkinson’s disease patients. The results also suggest that the low-frequency region of the Mel-spectrogram is more influential and important than the high-frequency region for Parkinson’s disease detection from speech.

12

Diagnosis of Parkinson’s disease based on SHAP value feature selection

Liu Yuchun, Liu Zhihui, Luo Xue, Zhao Hongjingtian

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 3

856--869

EN

To address the problem of high feature dimensionality of Parkinson’s disease medical data, this paper introduces SHapley Additive exPlanations (SHAP) value for feature selection of Parkinson’s disease medical dataset. This paper combines SHAP value with four classifiers, namely deep forest (gcForest), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM) and random forest (RF), respectively. Then this paper applies them to Parkinson’s disease diagnosis. First, the classifier is used to calculate the magnitude of contribution of SHAP value to the features, then the features with significant contribution in the classification task are selected, and then the data after feature selection is used as input to classify the Parkinson’s disease dataset for diagnosis using the classifier. The experimental results show that compared to Fscore, analysis of variance (Anova-F) and mutual information (MI) feature selection methods, the four models based on SHAP-value feature selection achieved good classification results. The SHAP-gcForest model combined with gcForest achieves classification accuracy of 91.78% and F1-score of 0.945 when 150 features are selected. The SHAP-LightGBM model combined with LightGBM achieves classification accuracy and F1-score of 91.62% and 0.945 when 50 features are selected, respectively. The classification effectiveness is second only to the SHAP-gcForest model, but the SHAP-LightGBM model is more computationally efficient than the SHAP-gcForest model. Finally, the effectiveness of the proposed method is verified by comparing it with the results of existing literature. The findings demonstrate that machine learning with SHAP value feature selection method has good classification performance in the diagnosis of Parkinson’s disease, and provides a reference for physicians in the diagnosis and prevention of Parkinson’s disease.

13

An investigation about the relationship between dysarthria level of speech and the neurological state of Parkinson’s patients

Karan Biswajit, Sekhar Sahu Sitanshu, Orozco-Arroyave Juan Rafael

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 2

710--726

EN

Parkinson’s disease (PD) is the most common neurological disorder that typically affects elderly people. In the earlier stage of disease, it has been seen that 90% of the patients develop voice disorders namely hypokinetic dysarthria. As time passes, the severity of PD increases, and patients have difficulty performing different speech tasks. During the progression of the disease, due to less control of articulatory organs such as the tongue, jaw, and lips, the quality of speech signals deteriorates. Periodic medical evaluations are very important for PD patients; however, having access to a medical appointment with a neurologist is a privilege in most countries. Considering that the speech recording process is inexpensive and very easy to do, we want to explore in this paper the suitability of mapping information of the dysarthria level into the neurological state of patients and vice versa. Three levels of severity are considered in a multiclass framework using time-frequency (TF) features and random-forest along with an Error-Correcting Output Code (ECOC) approach. The multiclass classification task based on dysarthria level is performed using the TF features with words and diadochokinetic (DDK) speech tasks. The developed model shows an unweighted average recall (UAR) of 68.49% with the DDK task /pakata/ based on m-FDA level, and 48.8% with the word /petaka/ based on the UPDRS level using the Random Forest classifier. With the aim, to evaluate the neurological states using the dysarthria level, the developed models are used to predict the MDS-UPDRS-III level of patients. The highest matching accuracy of 32% with the word /petaka/ is achieved. Similarly, the multiclass classification framework based on MDS-UPDRS-III is applied to predict the dysarthria level of patients. In this case, the highest matching accuracy of 18% was obtained with the DDK tasks /pataka/.

14

A hybrid two-stage SqueezeNet and support vector machine system for Parkinson’s disease detection based on handwritten spiral patterns

Bernardo Lucas Salvador, Damaševičius Robertas, de Albuquerque Victor Hugo C., Maskeliūnas Rytis

International Journal of Applied Mathematics and Computer Science

|

2021

|

Vol. 31, no. 4

549--561

EN

Parkinson’s disease (PD) is the second most common neurological disorder in the world. Nowadays, it is estimated that it affects from 2% to 3% of the global population over 65 years old. In clinical environments, a spiral drawing task is performed to help to obtain the disease’s diagnosis. The spiral trajectory differs between people with PD and healthy ones. This paper aims to analyze differences between handmade drawings of PD patients and healthy subjects by applying the SqueezeNet convolutional neural network (CNN) model as a feature extractor, and a support vector machine (SVM) as a classifier. The dataset used for training and testing consists of 514 handwritten draws of Archimedes’ spiral images derived from heterogeneous sources (digital and paper-based), from which 296 correspond to PD patients and 218 to healthy subjects. To extract features using the proposed CNN, a model is trained and 20% of its data is used for testing. Feature extraction results in 512 features, which are used for SVM training and testing, while the performance is compared with that of other machine learning classifiers such as a Gaussian naive Bayes (GNB) classifier (82.61%) and a random forest (RF) (87.38%). The proposed method displays an accuracy of 91.26%, which represents an improvement when compared to pure CNN-based models such as SqueezeNet (85.29%), VGG11 (87.25%), and ResNet (89.22%).

15

Deep learning vs feature engineering in the assessment of voice signals for diagnosis in Parkinson’s disease

Majda-Zdancewicz Ewelina, Potulska-Chromik Anna, Jakubowski Jacek, Nojszewska Monika, Kostera-Pruszczyk Anna

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2021

|

Vol. 69, nr 3

art. no. e137347

EN

Voice acoustic analysis can be a valuable and objective tool supporting the diagnosis of many neurodegenerative diseases, especially in times of distant medical examination during the pandemic. The article compares the application of selected signal processing methods and machine learning algorithms for the taxonomy of acquired speech signals representing the vowel a with prolonged phonation in patients with Parkinson’s disease and healthy subjects. The study was conducted using three different feature engineering techniques for the generation of speech signal features as well as the deep learning approach based on the processing of images involving spectrograms of different time and frequency resolutions. The research utilized real recordings acquired in the Department of Neurology at the Medical University of Warsaw, Poland. The discriminatory ability of feature vectors was evaluated using the SVM technique. The spectrograms were processed by the popular AlexNet convolutional neural network adopted to the binary classification task according to the strategy of transfer learning. The results of numerical experiments have shown different efficiencies of the examined approaches; however, the sensitivity of the best test based on the selected features proposed with respect to biological grounds of voice articulation reached the value of 97% with the specificity no worse than 93%. The results could be further slightly improved thanks to the combination of the selected deep learning and feature engineering algorithms in one stacked ensemble model.

16

Parkinson disease detection using energy direction features based on EMD from voice signal

Zhang Tao, Zhang Yajuan, Sun Hao, Shan Haoran

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 1

127--141

EN

Voice disorders are one of the incipient symptoms of Parkinson's disease (PD). Recent studies have shown that approximately 90% of PD patients suffer from vocal disorders. Therefore, it is significant to extract pathological information on the voice signals to detect PD. In this paper, a feature, named energy direction features based on empirical mode decomposition (EDF-EMD), is proposed to show the different characteristics of voice signals between PD patients and healthy subjects. Firstly, the intrinsic mode functions (IMFs) were obtained through the decomposition of voice signals by EMD. Then, the EDF is obtained by calculating the directional derivatives of the energy spectrum of each IMFs. Finally, the performance of the proposed feature is verified on two different datasets: dataset-Sakar and dataset-CPPDD. The proposed approach shows the best average resulting accuracy of 96.54% on dataset-Sakar and 92.59% on dataset-CPPDD. The results demonstrate that the method proposed in this paper is promising in the field of PD detection.

17

Detection of subthalamic nucleus using novel higher-order spectra features in microelectrode recordings signals

Hosny Mohamed, Zhu Minwei, Gao Wenpeng, Fu Yili

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 2

704--716

EN

Successful deep brain stimulation surgery for Parkinson’s disease (PD) patients hinges on accurate clustering of the functional regions along the electrode insertion trajectory. Microelectrode recording (MER) is employed as a substantial tool for neuro-navigation and localizing the optimal target, such as the subthalamic nucleus (STN), intraoperatively. MER signals deliver a framework to reveal the underlying characteristics of STN. The motivation behind this work is to explore the application of Higher-order statistics and spectra (HOS) for an automated delineation of the neurophysiological borders of STN using MER signals. Database collected from 21 PD patients were used. Two HOS methods (Bispectrum and cumulant) were exploited to probe non-Gaussian properties of STN region. This is followed by utilizing classifiers, namely K-nearest neighbor, decision tree, Boosting and support vector machine (SVM), to choose the superior classifier. Comparison of the performance achieved via HOS alongside the state-of-the-art techniques shows that the proposed features are better suited for identifying STN borders and achieve higher results. Average classification accuracy, sensitivity, specificity, area under the curve and Youden’s J statistics of 94.81%, 96.73%, 92.15%, 0.9444% and 0.8888, respectively, were yielded using SVM with 8 bispectrum and 241 cumulants features. The proposed model can aid the neurosurgeon in STN detection.

18

Detection of Parkinson’s disease using automated tunable Q wavelet transform technique with EEG signals

Khare Smith K., Bajaj Varun, Acharya U. Rajendra

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 2

679--689

EN

Deep brain simulations play an important role to study physiological and neuronal behavior during Parkinson’s disease (PD). Electroencephalogram (EEG) signals may faithfully represent the changes that occur during PD in the brain. But manual analysis of EEG signals is tedious, and time consuming as these signals are complex, non-linear, and non-stationary nature. Therefore EEG signals are required to decompose into multiple subbands (SBs) to get detailed and representative information from it. Experimental selection of basis function for the decomposition may cause system degradation due to information loss and an increased number of misclassification. To address this, an automated tunable Q wavelet transform (A-TQWT) is proposed for automatic decomposition. A-TQWT extracts representative SBs for analysis and provides better reconstruction for the synthesis of EEG signals by automatically selecting the tuning parameters. Five features are extracted from the SBs and classified different machine learning techniques. EEG dataset of 16 healthy controls (HC) and 15 PD (ON and OFF medication) subjects obtained from ”openneuro” is used to develop the automated model. We have aimed to develop an automated model that effectively classify HC subjects from PD patients with and without medication. The proposed method yielded an accuracy of 96.13% and 97.65% while the area under the curve of 97% and 98.56% for the classification of HC vs PD OFF medication and HC vs PD ON medication using least square support vector machine, respectively.

19

An improved framework for Parkinson’s disease prediction using Variational Mode Decomposition-Hilbert spectrum of speech signal

Karan Biswajit, Sekhar Sahu Sitanshu

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 2

717--732

EN

Parkinson’s disease (PD) is a neuro-degenerative disease due to loss of brain cells, which produces dopamine. It is most common after Alzheimer’s disease specially seen in old age people. In the earlier stage of disease, it has been noticed that most of the people suffering from speech disorder. From last two decades many studies have been conducted for the analysis of vocal tremors in PD. This study explores the combined approach of Variational Mode Decomposition (VMD) and Hilbert spectrum analysis (HSA) to investigate the voice tremor of patients with PD. A new set of features Hilbert cepstral coefficients (HCCs) are proposed in this study. Proposed features are assessed using vowels and words of PC-GITA database. The effectiveness of HCC features is utilized to perform classification, and regression analysis for PD detection. The highest average classification accuracy up to 91% and 96% is obtained with vowel /a/ and word /apto/ respectively. Further the classification accuracy up to 82% is obtained with independent dataset, when tested with the optimized model developed using PC-GITA database. In dysarthria level prediction highest correlation up to 0.82 is obtained using vowel /a/ and 0.8 with word /petaka/. The outcomes of this study indicate that the proposed articulatory features are suitable and accurate for PD assessment.

20

A novel deep recurrent convolutional neural network for subthalamic nucleus localization using local field potential signals

Hosny Mohamed, Zhu Minwei, Su Yixian, Gao Wenpeng, Fu Yili

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 4

1561--1574

EN

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is a well-established interventional treatment for improving motor symptoms of patients suffering from Parkinson’s disease (PD). While STN is originally localized using imaging modalities, additional intraoperative guidance such as microelectrode recording (MER) is crucial to refine the final electrode trajectory. Analysis of MER by an experienced neurophysiologist maintains good clinical outcomes, although the procedure requires long duration and jeopardizes the safety of the surgery. Lately, local field potentials (LFP) investigation has inspired the emergence of adaptive DBS and revealed beneficial perception of PD mechanisms. Several studies confronting LFP analysis to detect the anatomical borders of STN, have focused on handcrafted feature engineering, which does not certainly capture delicate differences in LFP. This study gauges the ability of deep learning to exhibit valuable insight into the electrophysiological neural rhythms of STN using LFP. A recurrent convolutional neural network (CNN) strategy is presented, where local features are extracted from LFP signals via CNN, followed by recurrent layers to aggregate the best features for classification. The proposed model outperformed the state-of-the-art techniques, yielding highest average accuracy of 96.79%. This is the first study on the analysis of LFP signals to localize STN using deep recurrent CNN. The developed model has the potential to extract high level biomarkers regarding STN region, which would boost the neurosurgeon’s confidence in adjusting the trajectory intraoperatively for optimal lead implantation. LFP is a robust guidance tool and could be an alternative solution to the current scenario using MER.