Globally optimal breast mass segmentation from DCE-MRI using deep semantic segmentation as shape prior

Maicas, Gabriel; Carneiro, Gustavo; Bradley, Andrew P.

doi:10.1109/isbi.2017.7950525

Cited by 24 publications

(37 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite a suboptimal training dataset, which was limited in size and contained some heterogeneity of MRI parameters, the U‐net achieved comparable Dice coefficient performance to recently published methods . This segmentation accuracy is achieved in a realistic, clinically challenging dataset which comprises several breast cancer subtypes and presentations.…”

Section: Discussionmentioning

confidence: 79%

“…Despite a suboptimal training dataset, which was limited in size and contained some heterogeneity of MRI parameters, the U-net achieved comparable Dice coefficient performance to recently published methods. 9,20 This segmentation accuracy is achieved in a realistic, clinically challenging dataset which comprises several breast cancer subtypes and presentations. A consequence of the limited dataset size is the fact that the CNN, when trained using data from either R1 or R3, generally overfit the training data, yielding a higher Dice coefficient across the training dataset than the testing set in both instances.…”

Section: Discussionmentioning

confidence: 98%

See 1 more Smart Citation

Task‐based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis

Spuhler

Ding

Liu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

Purpose Radiomics allows for powerful data‐mining and feature extraction techniques to guide clinical decision making. Image segmentation is a necessary step in such pipelines and different techniques can significantly affect results. We demonstrate that a convolutional neural network (CNN) segmentation method performs comparably to expert manual segmentations in an established radiomics pipeline. Methods Using the manual regions of interest (ROIs) of an expert radiologist (R1), a CNN was trained to segment breast lesions from dynamic contrast‐enhanced MRI (DCE‐MRI). Following network training, we segmented lesions for the testing set of a previously established radiomics pipeline for predicting lymph node metastases using DCE‐MRI of breast cancer. Prediction accuracy of CNN segmentations relative to manual segmentations by R1 from the original study, a resident (R2), and another expert radiologist (R3) were determined. We then retrained the CNN and radiomics model using R3’s manual segmentations to determine the effects of different expert observers on end‐to‐end prediction. Results Using R1’s ROIs, the CNN achieved a mean Dice coefficient of 0.71 ± 0.16 in the testing set. When input to our previously published radiomics pipeline, these CNN segmentations achieved comparable prediction performance to R1’s manual ROIs, and superior performance to those of the other radiologists. Similar results were seen when training the CNN and radiomics model using R3’s ROIs. Conclusion A CNN architecture is able to provide DCE‐MRI breast lesion segmentations which are suitable for input to our radiomics model. Moreover, the previously established radiomics model and CNN can be accurately trained end‐to‐end using ground truth data provided by distinct experts.

show abstract

Section: Discussionmentioning

confidence: 79%

Section: Discussionmentioning

confidence: 98%

Task‐based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis

Spuhler

Ding

Liu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…Another limitation of this work involves the lack of cross-validation experiments. This decision is justified to allow a fair comparison with other works (Maicas et al, 2017a(Maicas et al, ,b, 2018aMcClymont et al, 2014) on the same dataset. Future work involves the improvement of the malignant lesion localization in post-hoc methodologies by designing a new method specifically for the small training set available.…”

Section: Limitations and Future Workmentioning

confidence: 99%

Pre and post-hoc diagnosis and interpretation of malignancy from breast DCE-MRI

Maicas

Bradley

Nascimento

et al. 2019

Medical Image Analysis

Self Cite

View full text Add to dashboard Cite

We propose a new method for breast cancer screening from DCE-MRI based on a post-hoc approach that is trained using weakly annotated data (i.e., labels are available only at the image level without any lesion delineation). Our proposed post-hoc method automatically diagnosis the whole volume and, for positive cases, it localizes the malignant lesions that led to such diagnosis. Conversely, traditional approaches follow a pre-hoc approach that initially localises suspicious areas that are subsequently classified to establish the breast malignancy -this approach is trained using strongly annotated data (i.e., it needs a delineation and classification of all lesions in an image). Another goal of this paper is to establish the advantages and disadvantages of both approaches when applied to breast screening from DCE-MRI. Relying on experiments on a breast DCE-MRI dataset that contains scans of 117 patients, our results show that the post-hoc method is more accurate for diagnosing the whole volume per patient, achieving an AUC of 0.91, while the pre-hoc method achieves an AUC of 0.81. However, the performance for localising the malignant lesions remains challenging for the post-hoc method due to the weakly labelled dataset employed during training.

show abstract

“…The idea behind their contribution is to model the underlying dynamics of the tumor by a linear dynamics system and use the system parameters to segment the cancer. Maicas et al [27] have proposed a method based on a globally optimal inference in a continuous space using a shape prior computed from a semantic segmentation produced by a deep learning model. Despite the fact that the presented state-of-the-art methods are based on strong theoretical foundations, they actually generate a large amount of computations per final output.…”

Section: Other Approachesmentioning

confidence: 99%

“…Nevertheless, extraction of breast tumor from MR images is a challenging task due to their considerable shape variations, intensity inhomogeneity, overlap with the normal breast tissue, etc. Several segmentation methods are proposed in the literature to address these issues including contour-based [11][12][13][14] and region-based [15][16][17][22][23][24][25][26][27] approaches. Unfortunately, none of the state-of-the-art methods produced rigorous results and the challenge remains open.…”

Section: Introductionmentioning

confidence: 99%

Levels Propagation Approach to Image Segmentation: Application to Breast MR Images

Bouchebbah

Slimani

2019

J Digit Imaging

View full text Add to dashboard Cite

Accurate segmentation of a breast tumor region is fundamental for treatment. Magnetic resonance imaging (MRI) is a widely used diagnostic tool. In this paper, a new semi-automatic segmentation approach for MRI breast tumor segmentation called Levels Propagation Approach (LPA) is introduced. The introduced segmentation approach takes inspiration from tumor propagation and relies on a finite set of nested and non-overlapped levels. LPA has several features: it is highly suitable to parallelization and offers a simple and dynamic possibility to automate the threshold selection. Furthermore, it allows stopping of the segmentation at any desired limit. Particularly, it allows to avoid to reach the breast skin-line region which is known as a significant issue that reduces the precision and the effectiveness of the breast tumor segmentation. The proposed approach have been tested on two clinical datasets, namely RIDER breast tumor dataset and CMH-LIMED breast tumor dataset. The experimental evaluations have shown that LPA has produced competitive results to some state-of-the-art methods and has acceptable computation complexity.

show abstract

Globally optimal breast mass segmentation from DCE-MRI using deep semantic segmentation as shape prior

Cited by 24 publications

References 18 publications

Task‐based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis

Task‐based assessment of a convolutional neural network for segmenting breast lesions for radiomic analysis

Pre and post-hoc diagnosis and interpretation of malignancy from breast DCE-MRI

Levels Propagation Approach to Image Segmentation: Application to Breast MR Images

Contact Info

Product

Resources

About