Szczegóły publikacji
Opis bibliograficzny
The influence of textural features on the differentiation of coronary vessel wall lesions visualized on IVUS images / Weronika Małek, Tomasz Roleder, Elżbieta POCIASK // W: Information Technology in Biomedicine : 9th international conference, ITIB 2022 : Kamień Śląski, Poland, June 20-22, 2022 : proceedings / eds. Ewa Pietka, [et al.]. — Cham: Springer International Publishing AG, cop. 2022. — (Advances in Intelligent Systems and Computing ; ISSN 2194-5357 ; vol. 1429). — ISBN: 978-3-031-09134-6; e-ISBN: 978-3-031-09135-3. — S. 181–193. — Bibliogr., Abstr.
Autorzy (3)
- AGHMałek Weronika
- Roleder Tomasz
- AGHPociask Elżbieta
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 145168 |
|---|---|
| Data dodania do BaDAP | 2023-02-09 |
| DOI | 10.1007/978-3-031-09135-3_16 |
| Rok publikacji | 2022 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Wydawca | Springer |
| Czasopismo/seria | Advances in Intelligent Systems and Computing |
Abstract
Distinguishing atherosclerotic plaque from other image elements is crucial for patient diagnosis and treatment. Due to the similar echogenicity of soft (lipid) plaques and blood, there is a possibility of misinterpreting the size of the vessel lumen, which is a key element in correct stent selection. In this study, by means of texture analysis of the IVUS image area in which the plaque is located, plaque was distinguished from the lumen of the vessel in which blood is present. qMaZda software was used for this as it allows extensive analysis of image texture. The analysis was performed for features that were derived from the brightness of the image. Subsequently, the focus was on statistical analysis of the obtained features. The texture features that allow differentiation between blood in the lumen area and plaque were determined using Pearson’s and Spearman’s correlation. In order to obtain good classification accuracy, classification of the values of the selected parameters was performed using logistic regression. The results were presented using a confusion matrix and a plot of logistic regression. The texture features that are capable of differentiating plaque from blood in IVUS images were identified.
