2019
DOI: 10.3348/kjr.2018.0530
|View full text |Cite
|
Sign up to set email alerts
|

Effect of a Deep Learning Framework-Based Computer-Aided Diagnosis System on the Diagnostic Performance of Radiologists in Differentiating between Malignant and Benign Masses on Breast Ultrasonography

Abstract: Objective To investigate whether a computer-aided diagnosis (CAD) system based on a deep learning framework (deep learning-based CAD) improves the diagnostic performance of radiologists in differentiating between malignant and benign masses on breast ultrasound (US). Materials and Methods B-mode US images were prospectively obtained for 253 breast masses (173 benign, 80 malignant) in 226 consecutive patients. Breast mass US findings were retrospectively analyzed by deep… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

7
67
2

Year Published

2020
2020
2022
2022

Publication Types

Select...
7
1
1

Relationship

0
9

Authors

Journals

citations
Cited by 91 publications
(76 citation statements)
references
References 35 publications
7
67
2
Order By: Relevance
“…Currently, the convolutional neural network (CNN), a deep learning technique, is widely used in the medical field due to its aid in reaching an accurate diagnosis, reducing medical errors, and improving productivity (11)(12)(13). Furthermore, CNNs have also been successfully employed in thoracic CT, for example, in the automated classification of pulmonary nodules (14,15), carcinoma, or tuberculosis (16).…”
Section: Dataset and Classification Criteriamentioning
confidence: 99%
“…Currently, the convolutional neural network (CNN), a deep learning technique, is widely used in the medical field due to its aid in reaching an accurate diagnosis, reducing medical errors, and improving productivity (11)(12)(13). Furthermore, CNNs have also been successfully employed in thoracic CT, for example, in the automated classification of pulmonary nodules (14,15), carcinoma, or tuberculosis (16).…”
Section: Dataset and Classification Criteriamentioning
confidence: 99%
“…Shibusawa et al reported that CAD could significantly increase the AUC of the observers from 0.649 to 0.783 (p = 0.0167) (12). A recent study showed that adding CAD results to ultrasound significantly improved the specificity, accuracy, and PPV of radiologists without losing sensitivity and NPV (17).…”
Section: Discussionmentioning
confidence: 99%
“…Thus, deep learningbased CAD is helpful in distinguishing benign from worrisome lesions. Choi et al (17) also found that deep learning-based CAD could improve the diagnostic performance of leading radiologists and enable radiologists to correctly diagnose lesions that are difficult to classify as BI-RADS 3 or 4a.…”
Section: For the Bi-rads 4a Lesions Of The Asymptomatic Patientsmentioning
confidence: 99%
“…Resampling Process [23] 2019 Breast tumor Histology Resampling Process [24] 2019 Breast tumor Histology Not mentioned [25] 2018 Liver fibrosis Histology Resampling Process [26] 2017 Lung cancer Expert consent…”
Section: Expert Consentmentioning
confidence: 99%