BIRADS features-oriented semi-supervised deep learning for breast ultrasound computer-aided diagnosis.

08:00 EDT 10th March 2020 | BioPortfolio

Summary of "BIRADS features-oriented semi-supervised deep learning for breast ultrasound computer-aided diagnosis."

We propose a novel BIRADS-SSDL network that integrates clinically-approved breast lesion characteristics (BIRADS features) into a task-oriented Semi-Supervised Deep Learning (SSDL) for accurate diagnosis on ultrasound (US) images with a small training dataset. Breast US images are converted to BIRADS-oriented Feature Maps (BFMs) using a distance-transformation coupled with a Gaussian filter. Then, the converted BFMs are used as the input of an SSDL network, which performs unsupervised Stacked Convolutional Auto-Encoder (SCAE) image reconstruction guided by lesion classification. This integrated multi-task learning allows SCAE to extract image features with the constraints from the lesion classification task, while the lesion classification is achieved by utilizing the SCAE encoder features with a convolutional network. We trained BIRADS-SSDL network with an alternative learning strategy by balancing reconstruction error and classification label prediction error. We compared the performance of the BIRADS-SSDL network with conventional SCAE and SSDL methods that use original images as inputs, as well as with an SCAE that use BFMs as inputs. Experimental results on two breast US datasets show that BIRADS-SSDL ranked the best among the four networks, with classification accuracy around 94.23±3.33% and 84.38±3.11%. In the case of experiments across two datasets collected from two different institution/and US devices, the developed BIRADS-SSDL is generalizable across the different US devices and institutions without overfitting to a single dataset and achieved satisfactory results. Furthermore, we investigate the performance of the proposed method by varying model training strategies, lesion boundary accuracy, and Gaussian filter parameter. Experimental results showed that pre-training strategy can help to speed up model convergence during training but no improvement of classification accuracy on testing dataset. The proposed method could achieve a satisfactory performance when there are slight deviation of the lesion's boundary. Compared with state-of-the-art methods, BIRADS-SSDL could be promising for effective breast US lesion CAD using small datasets.


Journal Details

This article was published in the following journal.

Name: Physics in medicine and biology
ISSN: 1361-6560


DeepDyve research library

PubMed Articles [15087 Associated PubMed Articles listed on BioPortfolio]

Chronic gastritis classification using gastric X-ray images with a semi-supervised learning method based on tri-training.

High-quality annotations for medical images are always costly and scarce. Many applications of deep learning in the field of medical image analysis face the problem of insufficient annotated data. In ...

Semi-Supervised Image Dehazing.

We present an effective semi-supervised learning algorithm for single image dehazing. The proposed algorithm applies a deep Convolutional Neural Network (CNN) containing a supervised learning branch a...

A Deep Supervised Learning Framework for Data-Driven Soft Sensor Modeling of Industrial Processes.

Deep learning has been recently introduced for soft sensors in industrial processes. However, most of the existing deep networks, such as stacked autoencoder, are pretrained in a layerwise unsupervise...

Semi-Supervised Graph Regularized Deep NMF With Bi-Orthogonal Constraints for Data Representation.

Semi-supervised non-negative matrix factorization (NMF) exploits the strengths of NMF in effectively learning local information contained in data and is also able to achieve effective learning when on...

Automatic Identification of Breast Ultrasound Image Based on Supervised Block-Based Region Segmentation Algorithm and Features Combination Migration Deep Learning Model.

Breast cancer is a high-incidence type of cancer for women. Early diagnosis plays a crucial role in the successful treatment of the disease and the effective reduction of deaths. In this paper, deep l...

Clinical Trials [11112 Associated Clinical Trials listed on BioPortfolio]

Automated Breast Ultrasound Screening

Women with dense breasts (BIRADS 3 or 4) with a BIRADS category 1 or 2 mammogram are asked to participate in a automated volume breast ultrasound scanner. The study evaluates prospectively...

Application of Deep-learning and Ultrasound Elastography in Opportunistic Screening of Breast Cancer

As the most common cancer expected to occur all over the world, breast cancer still faces with the unsatisfied diagnostic accuracy in US imaging. S-detect is a sophisticated CAD system for...

Efficacy of Tomosynthesis in the BIRADS 3 Population

There are three main objectives, centering on those patients targeted for surveillance imaging. The first objective is to compare image quality and specificity of breast tomosynthesis and...

OctavaPink for Women With Dense Breast After Negative Mammography

The OctavaPink test is indicated as an adjunct to mammography for breast cancer screening in asymptomatic women for whom screening or diagnostic mammography findings are normal or benign ...

ARFI, VisR and DDAI Ultrasound for Improving Discrimination of Malignant and Unresponsive Breast Cancer

Purpose: The purpose of this study is to evaluate in vivo the diagnostic relevance of ultrasound-derived metrics for stiffness, elasticity, viscosity, and anisotropy. Participants: Forty ...

Medical and Biotech [MESH] Definitions

A MACHINE LEARNING paradigm used to make predictions about future instances based on a given set of labeled paired input-output training (sample) data.

SUPERVISED MACHINE LEARNING algorithm which learns to assign labels to objects from a set of training examples. Examples are learning to recognize fraudulent credit card activity by examining hundreds or thousands of fraudulent and non-fraudulent credit card activity, or learning to make disease diagnosis or prognosis based on automatic classification of microarray gene expression profiles drawn from hundreds or thousands of samples.

Process in which individuals take the initiative, in diagnosing their learning needs, formulating learning goals, identifying resources for learning, choosing and implementing learning strategies and evaluating learning outcomes (Knowles, 1975)

A fluid-filled closed cavity or sac that is lined by an EPITHELIUM and found in the BREAST. It may appear as a single large cyst in one breast, multifocal, or bilateral in FIBROCYSTIC BREAST DISEASE.

Change in learning in one situation due to prior learning in another situation. The transfer can be positive (with second learning improved by first) or negative (where the reverse holds).

Quick Search

DeepDyve research library

Relevant Topics

Breast Cancer
Track and monitor developments in breast cancer research and commercial development.  Follow the tabs above to read the latest global news, research, clinical trials on breast cancer and follow companies active in the development of breast cancer tr...

Radiology is the branch of medicine that studies imaging of the body; X-ray (basic, angiography, barium swallows), ultrasound, MRI, CT and PET. These imaging techniques can be used to diagnose, but also to treat a range of conditions, by allowing visuali...

Antiretroviral therapy
Standard antiretroviral therapy (ART) consists of the combination of at least three antiretroviral (ARV) drugs to maximally suppress the HIV virus and stop the progression of HIV disease. Huge reductions have been seen in rates of death and suffering whe...

Searches Linking to this Article