scholarly journals Lesion Detection in Breast Tomosynthesis Using Efficient Deep Learning and Data Augmentation Techniques

2021 ◽  
Author(s):  
Loay Hassan ◽  
Mohamed Abedl-Nasser ◽  
Adel Saleh ◽  
Domenec Puig
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Breast Lesion ◽  
Data Augmentation ◽  
Digital Breast Tomosynthesis ◽  
Lesion Detection ◽  
Detection Methods ◽  
Breast Tomosynthesis ◽  
Mammographic Images ◽  
Augmentation Techniques

Digital breast tomosynthesis (DBT) is one of the powerful breast cancer screening technologies. DBT can improve the ability of radiologists to detect breast cancer, especially in the case of dense breasts, where it beats mammography. Although many automated methods were proposed to detect breast lesions in mammographic images, very few methods were proposed for DBT due to the unavailability of enough annotated DBT images for training object detectors. In this paper, we present fully automated deep-learning breast lesion detection methods. Specifically, we study the effectiveness of two data augmentation techniques (channel replication and channel-concatenation) with five state-of-the-art deep learning detection models. Our preliminary results on a challenging publically available DBT dataset showed that the channel-concatenation data augmentation technique can significantly improve the breast lesion detection results for deep learning-based breast lesion detectors.

scholarly journals Breast Cancer Conspicuity on Simultaneously Acquired Digital Mammographic Images versus Digital Breast Tomosynthesis Images

Radiology ◽  
2019 ◽  
Vol 292 (1) ◽  
pp. 69-76 ◽  
Author(s):  
Katrina E. Korhonen ◽  
Emily F. Conant ◽  
Eric A. Cohen ◽  
Marie Synnestvedt ◽  
Elizabeth S. McDonald ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Digital Breast Tomosynthesis ◽  
Breast Tomosynthesis ◽  
Mammographic Images

scholarly journals Preoperative Staging in Breast Cancer: Intraindividual Comparison of Unenhanced MRI Combined With Digital Breast Tomosynthesis and Dynamic Contrast Enhanced-MRI

2021 ◽  
Vol 11 ◽  
Author(s):  
Veronica Rizzo ◽  
Giuliana Moffa ◽  
Endi Kripa ◽  
Claudia Caramanico ◽  
Federica Pediconi ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Preoperative Staging ◽  
Digital Breast Tomosynthesis ◽  
Lesion Size ◽  
Lesion Detection ◽  
P Value ◽  
Breast Tomosynthesis ◽  
Dce Mri ◽  
Mcnemar Test ◽  
Adc Values

ObjectivesTo evaluate the accuracy in lesion detection and size assessment of Unenhanced Magnetic Resonance Imaging combined with Digital Breast Tomosynthesis (UE-MRI+DBT) and Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI), in women with known breast cancer.MethodsA retrospective analysis was performed on 84 patients with histological diagnosis of breast cancer, who underwent MRI on a 3T scanner and DBT over 2018-2019, in our Institution. Two radiologists, with 15 and 7 years of experience in breast imaging respectively, reviewed DCE-MRI and UE-MRI (including DWI and T2-w) + DBT images in separate reading sections, unaware of the final histological examination. DCE-MRI and UE-MRI+DBT sensitivity, positive predictive value (PPV) and accuracy were calculated, using histology as the gold standard. Spearman correlation and regression analyses were performed to evaluate lesion size agreement between DCE-MRI vs Histology, UE-MRI+DBT vs Histology, and DCE-MRI vs UE-MRI+DBT. Inter-reader agreement was evaluated using Cohen’s κ coefficient. McNemar test was used to identify differences in terms of detection rate between the two methodological approaches. Spearman’s correlation analysis was also performed to evaluate the correlation between ADC values and histological features.Results109 lesions were confirmed on histological examination. DCE-MRI showed high sensitivity (100% Reader 1, 98% Reader 2), good PPV (89% Reader 1, 90% Reader 2) and accuracy (90% for both readers). UE-MRI+DBT showed 97% sensitivity, 91% PPV and 92% accuracy, for both readers. Lesion size Spearman coefficient were 0.94 (Reader 1) and 0.91 (Reader 2) for DCE-MRI vs Histology; 0.91 (Reader 1) and 0.90 (Reader 2) for UE-MRI+DBT vs Histology (p-value <0.001). DCE-MRI vs UE-MRI+DBT regression coefficient was 0.96 for Reader 1 and 0.94 for Reader 2. Inter-reader agreement was 0.79 for DCE-MRI and 0.94 for UE-MRI+DBT. McNemar test did not show a statistically significant difference between DCE-MRI and UE-MRI+DBT (McNemar test p-value >0.05). Spearman analyses showed an inverse correlation between ADC values and histological grade (p-value <0.001).ConclusionsDCE-MRI was the most sensitive imaging technique in breast cancer preoperative staging. However, UE-MRI+DBT demonstrated good sensitivity and accuracy in lesion detection and tumor size assessment. Thus, UE-MRI could be a valid alternative when patients have already performed DBT.

scholarly journals A deep learning classifier for digital breast tomosynthesis

2021 ◽  
Vol 83 ◽  
pp. 184-193
Author(s):  
R. Ricciardi ◽  
G. Mettivier ◽  
M. Staffa ◽  
A. Sarno ◽  
G. Acampora ◽  
...  
Keyword(s):  
Deep Learning ◽  
Digital Breast Tomosynthesis ◽  
Breast Tomosynthesis ◽  
Learning Classifier

scholarly journals Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study

2021 ◽  
Author(s):  
Suzanne L. van Winkel ◽  
Alejandro Rodríguez-Ruiz ◽  
Linda Appelman ◽  
Albert Gubern-Mérida ◽  
Nico Karssemeijer ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Artificial Intelligence ◽  
Cancer Detection ◽  
Support System ◽  
Reading Time ◽  
Digital Breast Tomosynthesis ◽  
Breast Tomosynthesis ◽  
Screening Programs ◽  
The Impact

Abstract Objectives Digital breast tomosynthesis (DBT) increases sensitivity of mammography and is increasingly implemented in breast cancer screening. However, the large volume of images increases the risk of reading errors and reading time. This study aims to investigate whether the accuracy of breast radiologists reading wide-angle DBT increases with the aid of an artificial intelligence (AI) support system. Also, the impact on reading time was assessed and the stand-alone performance of the AI system in the detection of malignancies was compared to the average radiologist. Methods A multi-reader multi-case study was performed with 240 bilateral DBT exams (71 breasts with cancer lesions, 70 breasts with benign findings, 339 normal breasts). Exams were interpreted by 18 radiologists, with and without AI support, providing cancer suspicion scores per breast. Using AI support, radiologists were shown examination-based and region-based cancer likelihood scores. Area under the receiver operating characteristic curve (AUC) and reading time per exam were compared between reading conditions using mixed-models analysis of variance. Results On average, the AUC was higher using AI support (0.863 vs 0.833; p = 0.0025). Using AI support, reading time per DBT exam was reduced (p < 0.001) from 41 (95% CI = 39–42 s) to 36 s (95% CI = 35– 37 s). The AUC of the stand-alone AI system was non-inferior to the AUC of the average radiologist (+0.007, p = 0.8115). Conclusions Radiologists improved their cancer detection and reduced reading time when evaluating DBT examinations using an AI reading support system. Key Points • Radiologists improved their cancer detection accuracy in digital breast tomosynthesis (DBT) when using an AI system for support, while simultaneously reducing reading time. • The stand-alone breast cancer detection performance of an AI system is non-inferior to the average performance of radiologists for reading digital breast tomosynthesis exams. • The use of an AI support system could make advanced and more reliable imaging techniques more accessible and could allow for more cost-effective breast screening programs with DBT.

Sign in / Sign up

Export Citation Format

Share Document