Breast ultrasound lesions recognition: end-to-end deep learning approaches

Yap, Moi Hoon; Goyal, Manu; Osman, Fatima; Martí, Robert; Denton, Erika R. E.; Juette, Arne; Zwiggelaar, Reyer

doi:10.1117/1.jmi.6.1.011007

Cited by 33 publications

(27 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [24] and [26], both the feature extraction (AUC = 0.849) and fine-tuning (AUC = 0.895) approaches were used, and the fine-tuning approach exhibited better performance. These results justify the fact that almost all of the previous studies on transfer learning applied to breast ultrasound [24][25][26][27][28][29] used fine-tuning to achieve superior performance (AUC = 0.895). However, in the performance analysis, the above conclusion does not provide sufficient insights into drawing a clear conclusion, because different studies used different methods (see Section 2.5) in terms of pre-processing, which highly affected performance; others even used different performance analysis metrics [23][24][25][26][27][28][29].…”

Section: Feature Extracting Vs Fine-tuningsupporting

confidence: 64%

“…For example, the last layer in a network that has been trained for classification would be highly specific to that classification task [49]. If the model was trained to classify tumors, one unit would respond only to the images of a specific tumor [23][24][25][26][27][28]. Transferring all layers except the top layer is the most common type of transfer learning [17][18][19][20].…”

Section: Advantages Of Transfer Learningmentioning

confidence: 99%

“…Transfer learning facilitates the building of a solid machine learning model with comparatively smaller training data because the model is already trained [53]. This is especially valuable in medical image processing because most of the time, data annotating persons are required to create large labeled datasets [24][25][26][27][28][29]. Furthermore, training time is minimized because it can reduce the time required to train a new deep neural network from the beginning in the case of complex target task [48,49].…”

Section: Advantages Of Transfer Learningmentioning

confidence: 99%

“…The most common pre-training models used for transfer learning in breast ultrasound are the VGG19, VGG16, AlexNet, and InceptionV3 models; VGG is the most common, followed by AlexNet and Inception, which are the least common. A comparison of the different pre-training models is not useful to determine the pre-training model that is better than the others for transfer learning in breast ultrasound [23][24][25][26][27][28][29]. However, one study [26], showed that Inception V3 outperforms VGG19, where the authors evaluated the impact of the ultrasound image reconstruction method on breast lesion classification using a neural transfer learning.…”

Section: Pre-training Model and Datasetmentioning

confidence: 99%

“…Shortly after this, Yap et al [25] published their work, which proposed the use of deep neural learning methods for breast cancer detection; they studied three different methods-a patch-based LeNet approach, a U-Net model, and a transfer learning method-with a pre-trained fully convolutional network, AlexNet. Following these works, a large number of articles have been published in the area of applying transfer learning for breast ultrasound imaging [26][27][28][29].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Transfer Learning in Breast Cancer Diagnoses via Ultrasound Imaging

Ayana

Dese

Choe

2021

Cancers

View full text Add to dashboard Cite

Transfer learning is a machine learning approach that reuses a learning method developed for a task as the starting point for a model on a target task. The goal of transfer learning is to improve performance of target learners by transferring the knowledge contained in other (but related) source domains. As a result, the need for large numbers of target-domain data is lowered for constructing target learners. Due to this immense property, transfer learning techniques are frequently used in ultrasound breast cancer image analyses. In this review, we focus on transfer learning methods applied on ultrasound breast image classification and detection from the perspective of transfer learning approaches, pre-processing, pre-training models, and convolutional neural network (CNN) models. Finally, comparison of different works is carried out, and challenges—as well as outlooks—are discussed.

show abstract

Section: Feature Extracting Vs Fine-tuningsupporting

confidence: 64%

Section: Advantages Of Transfer Learningmentioning

confidence: 99%

Section: Advantages Of Transfer Learningmentioning

confidence: 99%

Section: Pre-training Model and Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Transfer Learning in Breast Cancer Diagnoses via Ultrasound Imaging

Ayana

Dese

Choe

2021

Cancers

View full text Add to dashboard Cite

show abstract

Fully automatic tumor segmentation of breast ultrasound images with deep learning

Zhang

Liao

Wang

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

Background Breast ultrasound (BUS) imaging is one of the most prevalent approaches for the detection of breast cancers. Tumor segmentation of BUS images can facilitate doctors in localizing tumors and is a necessary step for computer‐aided diagnosis systems. While the majority of clinical BUS scans are normal ones without tumors, segmentation approaches such as U‐Net often predict mass regions for these images. Such false‐positive problem becomes serious if a fully automatic artificial intelligence system is used for routine screening. Methods In this study, we proposed a novel model which is more suitable for routine BUS screening. The model contains a classification branch that determines whether the image is normal or with tumors, and a segmentation branch that outlines tumors. Two branches share the same encoder network. We also built a new dataset that contains 1600 BUS images from 625 patients for training and a testing dataset with 130 images from 120 patients for testing. The dataset is the largest one with pixel‐wise masks manually segmented by experienced radiologists. Our code is available at https://github.com/szhangNJU/BUS_segmentation . Results The area under the receiver operating characteristic curve (AUC) for classifying images into normal/abnormal categories was 0.991. The dice similarity coefficient (DSC) for segmentation of mass regions was 0.898, better than the state‐of‐the‐art models. Testing on an external dataset gave a similar performance, demonstrating a good transferability of our model. Moreover, we simulated the use of the model in actual clinic practice by processing videos recorded during BUS scans; the model gave very low false‐positive predictions on normal images without sacrificing sensitivities for images with tumors. Conclusions Our model achieved better segmentation performance than the state‐of‐the‐art models and showed a good transferability on an external test set. The proposed deep learning architecture holds potential for use in fully automatic BUS health screening.

show abstract

Breast ultrasound image segmentation: A coarse‐to‐fine fusion convolutional neural network

et al. 2021

View full text Add to dashboard Cite

Purpose Breast ultrasound (BUS) image segmentation plays a crucial role in computer‐aided diagnosis systems for BUS examination, which are useful for improved accuracy of breast cancer diagnosis. However, such performance remains a challenging task owing to the poor image quality and large variations in the sizes, shapes, and locations of breast lesions. In this paper, we propose a new convolutional neural network with coarse‐to‐fine feature fusion to address the aforementioned challenges. Methods The proposed fusion network consists of an encoder path, a decoder path, and a core fusion stream path (FSP). The encoder path is used to capture the context information, and the decoder path is used for localization prediction. The FSP is designed to generate beneficial aggregate feature representations (i.e., various‐sized lesion features, aggregated coarse‐to‐fine information, and high‐resolution edge characteristics) from the encoder and decoder paths, which are eventually used for accurate breast lesion segmentation. To better retain the boundary information and alleviate the effect of image noise, we input the superpixel image along with the original image to the fusion network. Furthermore, a weighted‐balanced loss function was designed to address the problem of lesion regions having different sizes. We then conducted exhaustive experiments on three public BUS datasets to evaluate the proposed network. Results The proposed method outperformed state‐of‐the‐art (SOTA) segmentation methods on the three public BUS datasets, with average dice similarity coefficients of 84.71(±1.07), 83.76(±0.83), and 86.52(±1.52), average intersection‐over‐union values of 76.34(±1.50), 75.70(±0.98), and 77.86(±2.07), average sensitivities of 86.66(±1.82), 85.21(±1.98), and 87.21(±2.51), average specificities of 97.92(±0.46), 98.57(±0.19), and 99.42(±0.21), and average accuracies of 95.89(±0.57), 97.17(±0.3), and 98.51(±0.3). Conclusions The proposed fusion network could effectively segment lesions from BUS images, thereby presenting a new feature fusion strategy to handle challenging task of segmentation, while outperforming the SOTA segmentation methods. The code is publicly available at https://github.com/mniwk/CF2-NET.

show abstract

Breast ultrasound lesions recognition: end-to-end deep learning approaches

Cited by 33 publications

References 43 publications

Transfer Learning in Breast Cancer Diagnoses via Ultrasound Imaging

Transfer Learning in Breast Cancer Diagnoses via Ultrasound Imaging

Fully automatic tumor segmentation of breast ultrasound images with deep learning

Breast ultrasound image segmentation: A coarse‐to‐fine fusion convolutional neural network

Contact Info

Product

Resources

About