Machine Learning-Aided Monitoring and Prediction of Respiratory and Neurodegenerative Diseases Using Wearables
Loading...
Files
Date
Authors
ORCID
Advisor
Referee
Mark
P
Journal Title
Journal ISSN
Volume Title
Publisher
Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií
Abstract
This thesis focuses on wearables for health status monitoring, covering applications aimed at emergency solutions to the COVID-19 pandemic and aging society. The methods of ambient assisted living (AAL) are presented for the neurodegenerative disease Parkinson's disease (PD), facilitating 'aging in place' thanks to machine learning and around wearables - solutions of mHealth. Furthermore, the approaches using machine learning and wearables are discussed for early-stage COVID-19 detection. Firstly, a publicly available dataset containing COVID-19, influenza, and healthy control data was reused for research purposes. The solution presented in this thesis is considering the classification problem and outperformed the state-of-the-art methods, whereas the original paper introduced just anomaly detection. The proposed model in the thesis for early detection of COVID-19 achieved 78 % for the k-NN classifier. Moreover, a second dataset available on request was utilized for recognition between COVID-19 cases and two types of influenza. The scrutinisation in the form of the classification between the COVID-19 and Influensa groups is proposed as the extension to the research presented in the original paper. The accuracy of the distinction between COVID-19 cases and influenza in the middle of the pandemic was equal to 73 % thanks to the k-NN. Furthermore, the contribution as the classification model of two aforementioned combined datasets was provided, and COVID-19 cases were able to be distinguished from healthy controls with 73 % accuracy thanks to XGBoost algorithm. The undeniable advantage of the illustrated approaches is taking into consideration the incubation period and contagiousness of the disease. In addition, some solutions for the detection of the aforementioned aging society phenomenon are presented. This study explores the possibility of fusing computerised analysis of hypomimia and hypokinetic dysarthria for the spectrum of Czech speech exercises. The introduced dataset is unique in this field because of its diversity and myriad of speech exercises. The aim is to introduce a new techniques of PD diagnosis that could be easily integrated into mHealth systems. A classifier based on XGBoost was used, and SHAP values were used to ensure interpretability. The presented interpretability allows for the identification of clinically valuable biomarkers. Moreover, the fusion of video and audio modalities increased the balanced accuracy to 83 %. This methodology pointed out the most indicative speech exercise – tongue twister from the clinical point of view. Furthermore, this work belongs to just a few studies which tackle the subject of utilising multimodality for PD and this approach was profitable in contrast with a single modality. Another study, presented in this thesis, investigated the possibility of detecting Parkinson's disease by observing changes in emotion expression during difficult-to-pronounce speech exercises. The obtained model with XGBoost achieved 69 % accuracy for a tongue twister. The usage of facial features, emotion recognition, and computational analysis of tongue twister was proved to be successful in PD detection, which is the key novelty and contribution of this study. Additionally, the unique overview of potential methodologies suitable for the detection of PD based on sleep disorders was depicted.
This thesis focuses on wearables for health status monitoring, covering applications aimed at emergency solutions to the COVID-19 pandemic and aging society. The methods of ambient assisted living (AAL) are presented for the neurodegenerative disease Parkinson's disease (PD), facilitating 'aging in place' thanks to machine learning and around wearables - solutions of mHealth. Furthermore, the approaches using machine learning and wearables are discussed for early-stage COVID-19 detection. Firstly, a publicly available dataset containing COVID-19, influenza, and healthy control data was reused for research purposes. The solution presented in this thesis is considering the classification problem and outperformed the state-of-the-art methods, whereas the original paper introduced just anomaly detection. The proposed model in the thesis for early detection of COVID-19 achieved 78 % for the k-NN classifier. Moreover, a second dataset available on request was utilized for recognition between COVID-19 cases and two types of influenza. The scrutinisation in the form of the classification between the COVID-19 and Influensa groups is proposed as the extension to the research presented in the original paper. The accuracy of the distinction between COVID-19 cases and influenza in the middle of the pandemic was equal to 73 % thanks to the k-NN. Furthermore, the contribution as the classification model of two aforementioned combined datasets was provided, and COVID-19 cases were able to be distinguished from healthy controls with 73 % accuracy thanks to XGBoost algorithm. The undeniable advantage of the illustrated approaches is taking into consideration the incubation period and contagiousness of the disease. In addition, some solutions for the detection of the aforementioned aging society phenomenon are presented. This study explores the possibility of fusing computerised analysis of hypomimia and hypokinetic dysarthria for the spectrum of Czech speech exercises. The introduced dataset is unique in this field because of its diversity and myriad of speech exercises. The aim is to introduce a new techniques of PD diagnosis that could be easily integrated into mHealth systems. A classifier based on XGBoost was used, and SHAP values were used to ensure interpretability. The presented interpretability allows for the identification of clinically valuable biomarkers. Moreover, the fusion of video and audio modalities increased the balanced accuracy to 83 %. This methodology pointed out the most indicative speech exercise – tongue twister from the clinical point of view. Furthermore, this work belongs to just a few studies which tackle the subject of utilising multimodality for PD and this approach was profitable in contrast with a single modality. Another study, presented in this thesis, investigated the possibility of detecting Parkinson's disease by observing changes in emotion expression during difficult-to-pronounce speech exercises. The obtained model with XGBoost achieved 69 % accuracy for a tongue twister. The usage of facial features, emotion recognition, and computational analysis of tongue twister was proved to be successful in PD detection, which is the key novelty and contribution of this study. Additionally, the unique overview of potential methodologies suitable for the detection of PD based on sleep disorders was depicted.
This thesis focuses on wearables for health status monitoring, covering applications aimed at emergency solutions to the COVID-19 pandemic and aging society. The methods of ambient assisted living (AAL) are presented for the neurodegenerative disease Parkinson's disease (PD), facilitating 'aging in place' thanks to machine learning and around wearables - solutions of mHealth. Furthermore, the approaches using machine learning and wearables are discussed for early-stage COVID-19 detection. Firstly, a publicly available dataset containing COVID-19, influenza, and healthy control data was reused for research purposes. The solution presented in this thesis is considering the classification problem and outperformed the state-of-the-art methods, whereas the original paper introduced just anomaly detection. The proposed model in the thesis for early detection of COVID-19 achieved 78 % for the k-NN classifier. Moreover, a second dataset available on request was utilized for recognition between COVID-19 cases and two types of influenza. The scrutinisation in the form of the classification between the COVID-19 and Influensa groups is proposed as the extension to the research presented in the original paper. The accuracy of the distinction between COVID-19 cases and influenza in the middle of the pandemic was equal to 73 % thanks to the k-NN. Furthermore, the contribution as the classification model of two aforementioned combined datasets was provided, and COVID-19 cases were able to be distinguished from healthy controls with 73 % accuracy thanks to XGBoost algorithm. The undeniable advantage of the illustrated approaches is taking into consideration the incubation period and contagiousness of the disease. In addition, some solutions for the detection of the aforementioned aging society phenomenon are presented. This study explores the possibility of fusing computerised analysis of hypomimia and hypokinetic dysarthria for the spectrum of Czech speech exercises. The introduced dataset is unique in this field because of its diversity and myriad of speech exercises. The aim is to introduce a new techniques of PD diagnosis that could be easily integrated into mHealth systems. A classifier based on XGBoost was used, and SHAP values were used to ensure interpretability. The presented interpretability allows for the identification of clinically valuable biomarkers. Moreover, the fusion of video and audio modalities increased the balanced accuracy to 83 %. This methodology pointed out the most indicative speech exercise – tongue twister from the clinical point of view. Furthermore, this work belongs to just a few studies which tackle the subject of utilising multimodality for PD and this approach was profitable in contrast with a single modality. Another study, presented in this thesis, investigated the possibility of detecting Parkinson's disease by observing changes in emotion expression during difficult-to-pronounce speech exercises. The obtained model with XGBoost achieved 69 % accuracy for a tongue twister. The usage of facial features, emotion recognition, and computational analysis of tongue twister was proved to be successful in PD detection, which is the key novelty and contribution of this study. Additionally, the unique overview of potential methodologies suitable for the detection of PD based on sleep disorders was depicted.
Description
Citation
SKIBIŃSKA, J. Machine Learning-Aided Monitoring and Prediction of Respiratory and Neurodegenerative Diseases Using Wearables [online]. Brno: Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. 2023.
Document type
Document version
Date of access to the full text
Language of document
en
Study field
bez specializace
Comittee
doc. Ing. Jiří Mekyska, Ph.D. (předseda)
prof. Ing. Jaroslav Koton, Ph.D. (člen)
prof. Ing. Jiří Mišurec, CSc. (člen)
prof. Ing. Zdeněk Smékal, CSc. (člen)
doc. Ing. Radim Burget, Ph.D. (člen)
prof. Elena - Simona Lohan (člen)
Prof. dr. Peter Peer (člen)
Prof. Marcos Faundez-Zanuy (člen)
Prof. Anna Esposito (člen)
doc. Ing. Jiří Hošek, Ph.D. (člen)
prof. Yevgeni Koucheryavy, Ph.D. (člen)
Date of acceptance
2023-12-04
Defence
Obhajoba práce probíhala hybridní formou, tj. na Ústavu telekomunikací FEKT VUT v Brně a online prostřednictvím MS Teams. Setkání zahájil předseda doc. Mekyska, který uvítal uchazeče, členy komise a přítomné hosty. Slovo bylo předáno M.Sc. Eng. Justyně Skibińské, která prezentovala motivaci, dosažené výsledky a perspektivy další vědeckovýzkumné činnosti v oblasti tématu disertace. Následovalo čtení posudků a diskuze s prof. dr. Peterem Peerem. Dále probíhala diskuse i s dalšími členy komise a s hosty. V detailu byly diskutovány hlavní přínosy disertační práce. Pozornost byla také věnována možnostem budoucího aplikačního využití v praxi. Byly položeny i otázky související s generalizací natrénovaných modelů, výzkumem spojeným s EEG a monitorováním spánku, problémem s velikostí databáze, možnostmi zlepšení výsledků, možnostmi augmentace dat, a výpočetní náročnosti. M.Sc. Eng. Justyna Skibińská na dotazy reagovala pohotově a prokázala svoji erudici v oboru. Kromě členů komise se veřejné části zúčastnili také 4 hosti. Celková doba veřejného vystoupení trvala téměř dvě hodiny. Následovalo uzavřené jednání členů komise, která tajně hlasovala.
The PhD defense took place in a hybrid form, i.e. at the Department of Telecommunications FEEC BUT in Brno and online via MS Teams. The meeting was opened by the chairman doc. Mekyska, who welcomed the applicants, committee members and guests. The floor was given to M.Sc. Eng. Justyna Skibińská, who presented the motivation, achieved results and perspectives of further scientific
research activities in the area of the dissertation topic. This was followed by the reading of reviews and a discussion with prof. dr. Peter Peer. There was also a discussion with other members of the committee and guests. The main benefits of the dissertation were discussed in detail. Attention was also paid to the possibilities of future application use in practice. Questions related to the generalization of trained models, research related to EEG and sleep monitoring, the problem with the size of the database, the possibilities of improving the results, the possibilities of data augmentation, and computational demands were also discussed. M.Sc. Eng. Justyna Skibińská responded promptly to questions and proved her erudition in the field. In addition to the members of the committee, 4 guests also participated in the public part. The public performance lasted almost two hours. This was followed by a private meeting of the members of the committee, who voted secretly.
Result of defence
práce byla úspěšně obhájena
Document licence
Standardní licenční smlouva - přístup k plnému textu bez omezení