2021
DOI: 10.1016/j.cmpb.2021.106369
|View full text |Cite
|
Sign up to set email alerts
|

Diagnosis and grading of vesicoureteral reflux on voiding cystourethrography images in children using a deep hybrid model

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
13
0
2

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
2

Relationship

3
6

Authors

Journals

citations
Cited by 24 publications
(16 citation statements)
references
References 32 publications
0
13
0
2
Order By: Relevance
“…In the study, confusion matrices obtained in different models were given separately and compared. A confusion matrix is a table often used to describe the performance of a classification model on a set of test data for which the actual values are known [29]. An example confusion matrix is given in Figure 4.…”
Section: Resultsmentioning
confidence: 99%
“…In the study, confusion matrices obtained in different models were given separately and compared. A confusion matrix is a table often used to describe the performance of a classification model on a set of test data for which the actual values are known [29]. An example confusion matrix is given in Figure 4.…”
Section: Resultsmentioning
confidence: 99%
“…In a CNN-based trial of cystourethrography on pediatric vesicoureteral reflux (1228 images), features were extracted by hybridization of Googlenet, MobilenetV2, and Densenet201 models and concatenated. They then were optimized by mPMR and classified by SVM and KNN and showed very high accuracy of vesicoureteral reflux grades (normal to V) ranging from 95.5 to 96.9% 32 . Regarding the relatively low accuracy in the current trial, the recent hybrid DL models might be a potential solution to increase the accuracy.…”
Section: Discussionmentioning
confidence: 99%
“…Moreover, combined with positron emission tomography (PET)/CT, DL methods can automatically detect a small nodule within the lungs, even those no less than 2 cm in diameter, with 66.7% sensitivity and 84.5% specificity 31 . Recently, more advanced DL models, called hybrid methods, have been proposed; for example, hybridization of three DL-based algorithms of concatenation, optimization by minimum redundancy maximum relevance (mPMR), and classification by machine learning classifiers showed very high diagnostic yields (accuracy no less than 96.9%) 32 , 33 .…”
Section: Introductionmentioning
confidence: 99%
“…Çalışmada kullanılan başarım ölçüm matrisleri karışıklık matris kullanılarak hesaplanmaktadır. Yapılan çalışmada modellerin başarımlarını değerlendirmek için doğruluk, duyarlılık, özgünlük ve f_ölçütü değerlendirme ölçütleri kullanılmıştır [22].…”
Section: Tablo 2 Karmaşıklık Matrisiunclassified