Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
A Wavelet Transform-Based Neural Network Denoising Algorithm for Mobile Phonocardiography
Blekinge Institute of Technology, Faculty of Engineering, Department of Applied Signal Processing.
Univ Pavia, ITA.
Blekinge Institute of Technology, Faculty of Engineering, Department of Applied Signal Processing.
Blekinge Institute of Technology, Faculty of Engineering, Department of Applied Signal Processing.
2019 (English)In: Sensors, E-ISSN 1424-8220, Vol. 19, no 4, article id 957Article in journal (Refereed) Published
Abstract [en]

Cardiovascular pathologies cause 23.5% of human deaths, worldwide. An auto-diagnostic system monitoring heart activity, which can identify the early symptoms of cardiac illnesses, might reduce the death rate caused by these problems. Phonocardiography (PCG) is one of the possible techniques able to detect heart problems. Nevertheless, acoustic signal enhancement is required since it is exposed to various disturbances coming from different sources. The most common denoising enhancement is based on the Wavelet Transform (WT). However, the WT is highly susceptible to variations in the noise frequency distribution. This paper proposes a new adaptive denoising algorithm, which combines WT and Time Delay Neural Networks (TDNN). The acquired signal is decomposed by means of the WT using the coif five-wavelet basis at the tenth decomposition level and then provided as input to the TDNN. Besides the advantage of adaptive thresholding, the reason for using TDNNs is their capacity of estimating the Inverse Wavelet Transform (IWT). The best parameters of the TDNN were found for a NN consisting of 25 neurons in the first and 15 in the second layer and the delay block of 12 samples. The method was evaluated on several pathological heart sounds and on signals recorded in a noisy environment. The performance of the developed system with respect to other wavelet-based denoising approaches was validated by the online questionnaire.

Place, publisher, year, edition, pages
MDPI , 2019. Vol. 19, no 4, article id 957
Keywords [en]
adaptive filters, auscultation techniques, auto-diagnostic system, cardiovascular pathologies, Inverse Wavelet Transform (IWT), noise cancellation, signal denoising, Time Delay Neural Networks (TDNN)
National Category
Signal Processing
Identifiers
URN: urn:nbn:se:bth-17760DOI: 10.3390/s19040957ISI: 000460829200208PubMedID: 30813479OAI: oai:DiVA.org:bth-17760DiVA, id: diva2:1302229
Note

open access

Available from: 2019-04-04 Created: 2019-04-04 Last updated: 2022-02-10Bibliographically approved
In thesis
1. Sensors and Algorithms in Industry 4.0: Security and Health Preservation Applications
Open this publication in new window or tab >>Sensors and Algorithms in Industry 4.0: Security and Health Preservation Applications
2021 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Globalisation and technological digitisation have triggered an Industry 4.0. revolution.  The core of this revolution is autonomisation of complex processes, which require expert knowledge. The technical foundations of Industry 4.0 are IoT, Big Data and AI technologies. Nowadays, autonomous systems are widely used to increase human and environmental safety and to prevent health degradation.  Such non-industrial, life related applications demand high reliability as well as precision and accuracy, which challenge engineering science. 

The thesis objective is to provide suitable solutions for non-invasive, automated, and autonomous systems used for life protection and health maintenance. The proposed solutions enable non-invasive measurements by means of vision and acoustic sensors. The presented methods and systems are designed based on an analytical assessment of existing technologies and algorithms. New hardware solutions, signal and data processing methods, as well as classification and decision-making algorithms are proposed. Where required, additional customisations and modifications are applied. The systems and methods presented have been modelled and rigorously validated, and subsequently implemented and verified in a real environment.  

The scope of the thesis includes the assessment of functional requirements, precision, accuracy and reliability of life-related technological systems. It covers an analytical evaluation of proposed methods and algorithms of filtration, feature extraction, also detection, localization, identification, and classification of objects. The application fields are health monitoring, nature observation and facilitating collaborative frameworks in modern factories. 

The thesis specifically focuses on methods and algorithms of autonomous decision making concerning the risk of heart disease, the threat of fatal collision of rare birds with man-made structures and the prevention of accidents in modern robotised factories. It also deals with the implementation of the Industry 4.0 fundamentals, which are smart sensing, IoT and AI methods optimised to improve the system performance in a broad sense. The applied distributed computing method and machine-to-machine communication are aimed at limiting the data stream at an early stage of the decision-making process, and thus ensure the system’s cost-effectiveness. From the thesis, one can understand how the Industry 4.0 paradigm can contribute to autonomisation of compound processes and to increase system performance, without compromising its affordability.

The thesis is divided into two parts. The first, Prolegomena provides an overview of the sensors and algorithms applicable to industrial safety along with human health and nature preservation. This part also visualizes the relationships and interactions among the articles comprising the second part named Papers. In general, each of the enclosed six papers deals with the problem of autonomisation of complex processes in real-time and in a regular environment.

Place, publisher, year, edition, pages
Karlskrona: Blekinge Tekniska Högskola, 2021
Series
Blekinge Institute of Technology Doctoral Dissertation Series, ISSN 1653-2090 ; 4
Keywords
Acoustic Sensor, Artificial Intelligence, Autonomisation, Classification, Detection, Feature Extraction, Health Preservation, Identification, Internet of Things, Machine Learning, Multi-Sensor System, Safety System, Vision System
National Category
Engineering and Technology Signal Processing Computer Systems
Research subject
Systems Engineering
Identifiers
urn:nbn:se:bth-21387 (URN)978-91-7295-424-3 (ISBN)
Public defence
2021-09-10, Zoom, Campus Gräsvik, Karlskrona, 13:15 (English)
Opponent
Supervisors
Available from: 2021-05-11 Created: 2021-05-10 Last updated: 2021-09-20Bibliographically approved

Open Access in DiVA

fulltext(3421 kB)482 downloads
File information
File name FULLTEXT01.pdfFile size 3421 kBChecksum SHA-512
621f4d6c06294c8c0d60798aa350369532d72ff6b2d59d61be6dce8463f4ffbba158c7fe13b1165f4ac51445f9752150c573a527166f4b81581f5cb7c655de64
Type fulltextMimetype application/pdf

Other links

Publisher's full textPubMed

Authority records

Gradolewski, DawidJohansson, SvenKulesza, Wlodek

Search in DiVA

By author/editor
Gradolewski, DawidJohansson, SvenKulesza, Wlodek
By organisation
Department of Applied Signal Processing
In the same journal
Sensors
Signal Processing

Search outside of DiVA

GoogleGoogle Scholar
Total: 482 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
pubmed
urn-nbn

Altmetric score

doi
pubmed
urn-nbn
Total: 362 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf