期刊
JOURNAL OF BIOMEDICAL INFORMATICS
卷 113, 期 -, 页码 -出版社
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.jbi.2020.103656
关键词
Mammography; Computer assisted diagnosis; Deep learning; Segmentation
资金
- DARPA [FA86501827865, FA86501827882]
- NIH [U54EB020405, U01-CA242879]
- NSF [CCF1763315, CCF1563078, 1937301]
- ONR [N000141712266]
- National Cancer Institute
- Moore Foundation
- NXP
- Xilinx
- LETI-CEA
- Intel
- IBM
- Microsoft
- NEC
- Toshiba
- TSMC
- ARM
- Hitachi
- BASF
- Accenture
- Ericsson
- Qualcomm
- Analog Devices
- Okawa Foundation
- American Family Insurance
- Google Cloud
- Swiss Re
- Intelligence Community Postdoctoral Fellowship
- Stanford Human-Centered Artificial Intelligence Seed Grants Program
- Teradat
- Ant Financial
- VMWare
- Infosys
This study compared machine learning methods for classifying mass lesions on mammography images and found that a common segmentation-free CNN model substantially outperforms other methods. This indicates that representation learning techniques are advantageous for mammogram analysis.
Purpose: To compare machine learning methods for classifying mass lesions on mammography images that use predefined image features computed over lesion segmentations to those that leverage segmentation-free representation learning on a standard, public evaluation dataset. Methods: We apply several classification algorithms to the public Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM), in which each image contains a mass lesion. Segmentationfree representation learning techniques for classifying lesions as benign or malignant include both a Bag-ofVisual-Words (BoVW) method and a Convolutional Neural Network (CNN). We compare classification performance of these techniques to that obtained using two different segmentation-dependent approaches from the literature that rely on specific combinations of end classifiers (e.g. linear discriminant analysis, neural networks) and predefined features computed over the lesion segmentation (e.g. spiculation measure, morphological characteristics, intensity metrics). Results: We report area under the receiver operating characteristic curve (AZ) values for malignancy classification on CBIS-DDSM for each technique. We find average AZ values of 0.73 for a segmentation-free BoVW method, 0.86 for a segmentation-free CNN method, 0.75 for a segmentation-dependent linear discriminant analysis of Rubber-Band Straightening Transform features, and 0.58 for a hybrid rule-based neural network classification using a small number of hand-designed features. Conclusions: We find that malignancy classification performance on the CBIS-DDSM dataset using segmentationfree BoVW features is comparable to that of the best segmentation-dependent methods we study, but also observe that a common segmentation-free CNN model substantially and significantly outperforms each of these (p < 0.05). These results reinforce recent findings suggesting that representation learning techniques such as BoVW and CNNs are advantageous for mammogram analysis because they do not require lesion segmentation, the quality and specific characteristics of which can vary substantially across datasets. We further observe that segmentation-dependent methods achieve performance levels on CBIS-DDSM inferior to those achieved on the original evaluation datasets reported in the literature. Each of these findings reinforces the need for standardization of datasets, segmentation techniques, and model implementations in performance assessments of automated classifiers for medical imaging.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据