Abstract
PurposeThe aim of this study was to develop and test a post-processing technique for detection and classification of lesions according to the BI-RADS atlas in automated breast ultrasound (ABUS) based on deep convolutional neural networks (dCNNs).Methods and materialsIn this retrospective study, 645 ABUS datasets from 113 patients were included; 55 patients had lesions classified as high malignancy probability. Lesions were categorized in BI-RADS 2 (no suspicion of malignancy), BI-RADS 3 (probability of malignancy < 3%), and BI-RADS 4/5 (probability of malignancy > 3%). A deep convolutional neural network was trained after data augmentation with images of lesions and normal breast tissue, and a sliding-window approach for lesion detection was implemented. The algorithm was applied to a test dataset containing 128 images and performance was compared with readings of 2 experienced radiologists.ResultsResults of calculations performed on single images showed accuracy of 79.7% and AUC of 0.91 [95% CI: 0.85–0.96] in categorization according to BI-RADS. Moderate agreement between dCNN and ground truth has been achieved (κ: 0.57 [95% CI: 0.50–0.64]) what is comparable with human readers. Analysis of whole dataset improved categorization accuracy to 90.9% and AUC of 0.91 [95% CI: 0.77–1.00], while achieving almost perfect agreement with ground truth (κ: 0.82 [95% CI: 0.69–0.95]), performing on par with human readers. Furthermore, the object localization technique allowed the detection of lesion position slice-wise.ConclusionsOur results show that a dCNN can be trained to detect and distinguish lesions in ABUS according to the BI-RADS classification with similar accuracy as experienced radiologists.Key Points• A deep convolutional neural network (dCNN) was trained for classification of ABUS lesions according to the BI-RADS atlas.• A sliding-window approach allows accurate automatic detection and classification of lesions in ABUS examinations.
Highlights
Breast cancer is one of the most common causes of cancer death in females, and mortality rates are increasing worldwide [1]
A deep convolutional neural network was trained for classification of automated breast ultrasound (ABUS) lesions according to the BI-RADS atlas
We were able to show that a deep convolutional neural network can be trained to classify lesions in ABUS according to the American College of Radiology (ACR) BI-RADS classification with accuracies higher than 95% in both the training and the validation datasets
Summary
Breast cancer is one of the most common causes of cancer death in females, and mortality rates are increasing worldwide [1]. 2.3 million women were diagnosed with breast cancer in 2020 alone, causing 685,000 deaths worldwide [2]. Dense breast tissue decreases the sensitivity of conventional mammography in cancer detection [7,8,9,10]. For women with dense breast tissue or women unwilling to undergo mammography, breast ultrasound may be a robust alternative modality, used as an adjunct to mammography or as an independent screening modality providing an increased accuracy for cancer detection [11, 12] but results are highly dependent on operators’ skill and experience, on top of that requiring an expert’s interpretation of results that may be highly subjective
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
