Dissertation/ Thesis
Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie ; Machine Learning Methods in Electrical Impedance Tomography Reconstruction
| Τίτλος: | Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie ; Machine Learning Methods in Electrical Impedance Tomography Reconstruction |
|---|---|
| Συγγραφείς: | Kouakouo Nomvussi, Serge Ayme |
| Συνεισφορές: | Mikulka, Jan, Kořínek, Radim, Dušek, Jan |
| Στοιχεία εκδότη: | Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií |
| Έτος έκδοσης: | 2025 |
| Συλλογή: | Brno University of Technology (VUT): Digital Library / Vysoké učení technické v Brně: Digitální knihovně |
| Θεματικοί όροι: | Artificial neural network, density-based algorithm, electrical impedance tomography, k-nearest neighbors, radial basis function neural networks |
| Περιγραφή: | iii Abstract Reconstructing clear and meaningful images from noisy or incomplete data is a funda- mental challenge in areas such as medical imaging, remote sensing, and computer vision. Tra- ditional methods like Total Variation and Gauss-Newton often fall short when confronted with complex shapes or high noise levels, leading to limited accuracy and loss of structural detail. This dissertation presents a new approach to image reconstruction using a Cascaded Ra- dial Basis Function Neural Network (CRBFNN). The method features a two-stage neural ar- chitecture. In the first subnetwork, DBSCAN clustering and K-Nearest Neighbors (KNN) are used for center and spread estimation, respectively, allowing the model to adapt to the underly- ing data structure. The second subnetwork applies a fixed spread to ensure stability and com- putational efficiency during the refinement of the final output. This design enables the network to respond adaptively to different noise patterns while preserving structural consistency in its predictions. Comprehensive experiments were carried out using simulated data affected by white Gaussian noise, impulsive noise, and contact noise. Across all conditions, the CRBFNN con- sistently demonstrated strong performance, achieving a Structural Similarity Index (SSIM) of up to 0.991, a Correlation Coefficient (CC) of 0.983, and a notably low training Mean Squared Error (MSE) of 0.00066. It also outperformed several modern techniques, including DenseNet, CWGAN-AM, and Enhanced CNN, both in accuracy and robustness. Beyond accuracy, the model offers practical advantages. Once trained, CRBFNN pro- duces high-resolution 2D conductivity maps in approximately 1.3 seconds using only CPU re- sources, making it suitable for real-time applications and integration into modular EIT systems. This research highlights CRBFNN as a reliable and efficient tool for image reconstruction under diverse and challenging conditions. Looking ahead, future work will aim to enhance computa- tional efficiency through hardware ... |
| Τύπος εγγράφου: | doctoral or postdoctoral thesis |
| Περιγραφή αρχείου: | application/pdf; application/octet-stream; text/html |
| Γλώσσα: | English |
| Relation: | KOUAKOUO NOMVUSSI, S. Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie [online]. Brno: Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. 2025.; 170765; https://hdl.handle.net/11012/255700 |
| Διαθεσιμότητα: | https://hdl.handle.net/11012/255700 |
| Rights: | Standardní licenční smlouva - přístup k plnému textu bez omezení |
| Αριθμός Καταχώρησης: | edsbas.A13AC7B5 |
| Βάση Δεδομένων: | BASE |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://hdl.handle.net/11012/255700# Name: EDS - BASE (ns324271) Category: fullText Text: View record from BASE |
|---|---|
| Header | DbId: edsbas DbLabel: BASE An: edsbas.A13AC7B5 RelevancyScore: 822 AccessLevel: 3 PubType: Dissertation/ Thesis PubTypeId: dissertation PreciseRelevancyScore: 821.935607910156 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie ; Machine Learning Methods in Electrical Impedance Tomography Reconstruction – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Kouakouo+Nomvussi%2C+Serge+Ayme%22">Kouakouo Nomvussi, Serge Ayme</searchLink> – Name: Author Label: Contributors Group: Au Data: Mikulka, Jan<br />Kořínek, Radim<br />Dušek, Jan – Name: Publisher Label: Publisher Information Group: PubInfo Data: Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií – Name: DatePubCY Label: Publication Year Group: Date Data: 2025 – Name: Subset Label: Collection Group: HoldingsInfo Data: Brno University of Technology (VUT): Digital Library / Vysoké učení technické v Brně: Digitální knihovně – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Artificial+neural+network%22">Artificial neural network</searchLink><br /><searchLink fieldCode="DE" term="%22density-based+algorithm%22">density-based algorithm</searchLink><br /><searchLink fieldCode="DE" term="%22electrical+impedance+tomography%22">electrical impedance tomography</searchLink><br /><searchLink fieldCode="DE" term="%22k-nearest+neighbors%22">k-nearest neighbors</searchLink><br /><searchLink fieldCode="DE" term="%22radial+basis+function+neural+networks%22">radial basis function neural networks</searchLink> – Name: Abstract Label: Description Group: Ab Data: iii Abstract Reconstructing clear and meaningful images from noisy or incomplete data is a funda- mental challenge in areas such as medical imaging, remote sensing, and computer vision. Tra- ditional methods like Total Variation and Gauss-Newton often fall short when confronted with complex shapes or high noise levels, leading to limited accuracy and loss of structural detail. This dissertation presents a new approach to image reconstruction using a Cascaded Ra- dial Basis Function Neural Network (CRBFNN). The method features a two-stage neural ar- chitecture. In the first subnetwork, DBSCAN clustering and K-Nearest Neighbors (KNN) are used for center and spread estimation, respectively, allowing the model to adapt to the underly- ing data structure. The second subnetwork applies a fixed spread to ensure stability and com- putational efficiency during the refinement of the final output. This design enables the network to respond adaptively to different noise patterns while preserving structural consistency in its predictions. Comprehensive experiments were carried out using simulated data affected by white Gaussian noise, impulsive noise, and contact noise. Across all conditions, the CRBFNN con- sistently demonstrated strong performance, achieving a Structural Similarity Index (SSIM) of up to 0.991, a Correlation Coefficient (CC) of 0.983, and a notably low training Mean Squared Error (MSE) of 0.00066. It also outperformed several modern techniques, including DenseNet, CWGAN-AM, and Enhanced CNN, both in accuracy and robustness. Beyond accuracy, the model offers practical advantages. Once trained, CRBFNN pro- duces high-resolution 2D conductivity maps in approximately 1.3 seconds using only CPU re- sources, making it suitable for real-time applications and integration into modular EIT systems. This research highlights CRBFNN as a reliable and efficient tool for image reconstruction under diverse and challenging conditions. Looking ahead, future work will aim to enhance computa- tional efficiency through hardware ... – Name: TypeDocument Label: Document Type Group: TypDoc Data: doctoral or postdoctoral thesis – Name: Format Label: File Description Group: SrcInfo Data: application/pdf; application/octet-stream; text/html – Name: Language Label: Language Group: Lang Data: English – Name: NoteTitleSource Label: Relation Group: SrcInfo Data: KOUAKOUO NOMVUSSI, S. Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie [online]. Brno: Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. 2025.; 170765; https://hdl.handle.net/11012/255700 – Name: URL Label: Availability Group: URL Data: https://hdl.handle.net/11012/255700 – Name: Copyright Label: Rights Group: Cpyrght Data: Standardní licenční smlouva - přístup k plnému textu bez omezení – Name: AN Label: Accession Number Group: ID Data: edsbas.A13AC7B5 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.A13AC7B5 |
| RecordInfo | BibRecord: BibEntity: Languages: – Text: English Subjects: – SubjectFull: Artificial neural network Type: general – SubjectFull: density-based algorithm Type: general – SubjectFull: electrical impedance tomography Type: general – SubjectFull: k-nearest neighbors Type: general – SubjectFull: radial basis function neural networks Type: general Titles: – TitleFull: Metody strojového učení v rekonstrukci dat elektrické impedanční tomografie ; Machine Learning Methods in Electrical Impedance Tomography Reconstruction Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Kouakouo Nomvussi, Serge Ayme – PersonEntity: Name: NameFull: Mikulka, Jan – PersonEntity: Name: NameFull: Kořínek, Radim – PersonEntity: Name: NameFull: Dušek, Jan IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2025 Identifiers: – Type: issn-locals Value: edsbas |
| ResultId | 1 |