Amniotic fluid classification based on volume and echogenicity using single deep pocket and texture feature

Ayu, Putu Desiana Wulaning and Hartati, Sri and Musdholifah, Aina and Nurdiati, Detty S. (2021) Amniotic fluid classification based on volume and echogenicity using single deep pocket and texture feature. ICIC Express Letters, 15 (7). 681 – 691. ISSN 1881803X

Full text not available from this repository. (Request a copy)

Abstract

The amniotic cavity contains fluid that serves as a cushion for the growing fetus and prevents its collision with the uterine wall. The amniotic fluid observation is a routine program performed by obstetricians to determine the growth of fetal health. However, in diagnose echogenicity, there are still differences in perceptions between obstetricians. Concerning these problems, this study proposes a feature extraction model and classification of amniotic fluid types into normal echogenic, normal clear, oligohydramnios echogenic, oligohydramnios clear, and polyhydramnios clear. The proposed feature extraction adopts obstetrician's knowledge or techniques in measuring volume using the Single Deep Pocket (SDP) method. Furthermore, it proposes a texture feature using First Order Statistical (FOS) and Gray Level Co-occurrence Matrices (GLCM) for echogenicity. In addition, the oversampling method was conducted due to the limited availability of data samples, while the experiments were performed using initial image data of 92 b-mode ultrasonography amniotic fluid. The classification stage used the SVM method, which was analyzed on three different kernels, namely RBF, polynomial, and sigmoid. The results showed that the proposed feature with RBF kernel can achieve an average value for an accuracy of 81.4, precision of 80.8, recall of 81.4, F-measure of 81, and ROC of 0.88. © 2021 ICIC International. All rights reserved.

Item Type: Article
Additional Information: Cited by: 3
Uncontrolled Keywords: Amniotic fluid, Echogenicity, Single deep pocket feature, Texture feature, Classification
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Mathematics and Natural Sciences > Computer Science & Electronics Department
Depositing User: Sri JUNANDI
Date Deposited: 29 Oct 2024 04:58
Last Modified: 29 Oct 2024 04:58
URI: https://ir.lib.ugm.ac.id/id/eprint/8500

Actions (login required)

View Item
View Item