Breast Tissue Segmentation Research Articles

Quantitative assessment of background parenchymal enhancement is associated with lifetime breast cancer risk in screening MRI

ObjectivesTo compare the quantitative background parenchymal enhancement (BPE) in women with different lifetime risks and BRCA mutation status of breast cancer using screening MRI.Materials and methodsThis study included screening MRI of 535 women divided into three groups based on lifetime risk: nonhigh-risk women, high-risk women without BRCA mutation, and BRCA1/2 mutation carriers. Six quantitative BPE measurements, including percent enhancement (PE) and signal enhancement ratio (SER), were calculated on DCE-MRI after segmentation of the whole breast and fibroglandular tissue (FGT). The associations between lifetime risk factors and BPE were analyzed via linear regression analysis. We adjusted for risk factors influencing BPE using propensity score matching (PSM) and compared the BPE between different groups. A two-sided Mann–Whitney U-test was used to compare the BPE with a threshold of 0.1 for multiple testing issue-adjusted p values.ResultsAge, BMI, menopausal status, and FGT level were significantly correlated with quantitative BPE based on the univariate and multivariable linear regression analyses. After adjusting for age, BMI, menopausal status, hormonal treatment history, and FGT level using PSM, significant differences were observed between high-risk non-BRCA and BRCA groups in PEFGT (11.5 vs. 8.0%, adjusted p = 0.018) and SERFGT (7.2 vs. 9.3%, adjusted p = 0.066).ConclusionQuantitative BPE varies in women with different lifetime breast cancer risks and BRCA mutation status. These differences may be due to the influence of multiple lifetime risk factors. Quantitative BPE differences remained between groups with and without BRCA mutations after adjusting for known risk factors associated with BPE.Clinical relevance statementBRCA germline mutations may be associated with quantitative background parenchymal enhancement, excluding the effects of known confounding factors. This finding can provide potential insights into the cancer pathophysiological mechanisms behind lifetime risk models.Key PointsExpanding understanding of breast cancer pathophysiology allows for improved risk stratification and optimized screening protocols.Quantitative BPE is significantly associated with lifetime risk factors and differs between BRCA mutation carriers and noncarriers.This research offers a possible understanding of the physiological mechanisms underlying quantitative BPE and BRCA germline mutations.

310 (PB-126) Poster - Breast tissue segmentation in MR images using deep-learning

310 (PB-126) Poster - Breast tissue segmentation in MR images using deep-learning

Numerical and physical modeling of breast cancer based on image fusion and artificial intelligence

PurposeThe key problem raised in the paper is the change in the position of the breast tumor due to magnetic resonance imaging examinations in the abdominal position relative to the supine position during the surgical procedure. Changing the position of the patient leads to significant deformation of the breast, which leads to the inability to indicate the location of the neoplastic lesion correctly.MethodsThis study outlines a methodological process for treating cancer patients. Pre-qualification assessments are conducted for magnetic resonance imaging (MRI), and 3D scans are taken in three positions: supine with arms raised, supine surgical position (SS), and standing. MRI and standard ultrasonography (USG) imaging are performed, and breast and cancer tissue are segmented from the MRI images. Finite element analysis is used to simulate tissue behavior in different positions, and an artificial neural network is trained to predict tumor dislocation. Based on the model, a 3D-printed breast with a highlighted tumor is manufactured. This computer-aided analysis is used to create a detailed surgical plan, and lumpectomy surgery is performed in the SS. In addition, the geometry of the tumor is presented to the medical staff as a 3D-printed element.ResultsBy utilizing a comprehensive range of techniques, including pre-qualification assessment, 3D scanning, MRI and USG imaging, segmentation of breast and cancer tissue, model analysis, image fusion, finite element analysis, artificial neural network training, and additive manufacturing, a detailed surgical plan can be created for performing lumpectomy surgery in the supine surgical position.ConclusionThe new approach developed for the pre-operative assessment and surgical planning of breast cancer patients has demonstrated significant potential for improving the accuracy and efficacy of surgical procedures. This procedure may also help the pathomorphological justification. Moreover, transparent 3D-printed breast models can benefit breast cancer operation assistance. The physical and computational models can help surgeons visualize the breast and the tumor more accurately and detailedly, allowing them to plan the surgery with greater precision and accuracy.

Multi-class semantic segmentation of breast tissues from MRI images using U-Net based on Haar wavelet pooling

MRI images used in breast cancer diagnosis are taken in a lying position and therefore are inappropriate for reconstructing the natural breast shape in a standing position. Some studies have proposed methods to present the breast shape in a standing position using an ordinary differential equation of the finite element method. However, it is difficult to obtain meaningful results because breast tissues have different elastic moduli. This study proposed a multi-class semantic segmentation method for breast tissues to reconstruct breast shapes using U-Net based on Haar wavelet pooling. First, a dataset was constructed by labeling the skin, fat, and fibro-glandular tissues and the background from MRI images taken in a lying position. Next, multi-class semantic segmentation was performed using U-Net based on Haar wavelet pooling to improve the segmentation accuracy for breast tissues. The U-Net effectively extracted breast tissue features while reducing image information loss in a subsampling stage using multiple sub-bands. In addition, the proposed network is robust to overfitting. The proposed network showed a mIOU of 87.48 for segmenting breast tissues. The proposed networks demonstrated high-accuracy segmentation for breast tissue with different elastic moduli to reconstruct the natural breast shape.

Multiclass Segmentation of Breast Tissue and Suspicious Findings: A Simulation-Based Study for the Development of Self-Steering Tomosynthesis.

In breast tomosynthesis, multiple low-dose projections are acquired in a single scanning direction over a limited angular range to produce cross-sectional planes through the breast for three-dimensional imaging interpretation. We built a next-generation tomosynthesis system capable of multidirectional source motion with the intent to customize scanning motions around "suspicious findings". Customized acquisitions can improve the image quality in areas that require increased scrutiny, such as breast cancers, architectural distortions, and dense clusters. In this paper, virtual clinical trial techniques were used to analyze whether a finding or area at high risk of masking cancers can be detected in a single low-dose projection and thus be used for motion planning. This represents a step towards customizing the subsequent low-dose projection acquisitions autonomously, guided by the first low-dose projection; we call this technique "self-steering tomosynthesis." A U-Net was used to classify the low-dose projections into "risk classes" in simulated breasts with soft-tissue lesions; class probabilities were modified using post hoc Dirichlet calibration (DC). DC improved the multiclass segmentation (Dice = 0.43 vs. 0.28 before DC) and significantly reduced false positives (FPs) from the class of the highest risk of masking (sensitivity = 81.3% at 2 FPs per image vs. 76.0%). This simulation-based study demonstrated the feasibility of identifying suspicious areas using a single low-dose projection for self-steering tomosynthesis.

Breast mass regions classification from mammograms using convolutional neural networks and transfer learning.

This study introduces a novel approach aimed at enhancing the quality of digital mammography images through pre-processing techniques, to improve breast cancer detection accuracy. The primary objective is to enhance image resolution, thus leading to more precise breast tissue segmentation and subsequent classification utilizing convolutional neural networks (CNNs). Three recognized public mammography databases: CBIS-DDSM, Mini-MIAS, and Inbreast were used as pre-processing data. Our statistical findings revealed that the EDSR method (PSNR = 39.05 dB/ SSIM = 0.90) consistently outperformed the visual quality of images when compared to SR-RDN (PSNR = 32.68 dB/SSIM = 0.82). Similarly, UNet demonstrated superior performance over SegNet, boasting an average Intersection over Union (IoU) of 0.862, an average Dice coefficient of 0.991, and an accuracy rate of 0.947 in Region of Interest (RoI) segmentation results. In conclusion, the ResNet model contributed to enhanced accuracy compared to conventional machine learning algorithms. However, it did not surpass state-of-the-art deep CNN-based classifiers, achieving an accuracy rate of 75%. Abbreviations: AUC: Area under curve; CAD: Computer aided system; CC: Cranio caudal; CNN: Convolutional neural network; DNN: Deep neural network; DDSM: Digital Database for Screening Mammography; DM: Digital mammography; DL: Deep learning; EDSR: Enhanced Deep Residual Network; E2E: End to End; ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks; ESPCN: Efficient sub-pixel convolutional neural network; GAN: Generative adversarial network; HR: High resolution; IoU: Intersection over Union; LR: Low resolution; MDSR: Multi-scale deep super-resolution; MLO: Mediolateral Oblique; PSNR: Peak signal to Noise Ratio; RoI: Region of interest; RDN: Residua Dense Network; RDB: Residual Dense Block; RNN: Recurrent Neural Network; ReLU: Rectified Linear Unit; SR-GAN: Super-Resolution Using a Generative Adversarial Network; SSIM: Structural Similarity Index Metric; SISR: Single image super resolution; SegNet: Segmentation Network; TP: True positive; TN: True negative; FP: False positive; FN: False negative; VGG: Visual geometric group; VDSR: Very Deep Network for SR

MSCDA: Multi-level semantic-guided contrast improves unsupervised domain adaptation for breast MRI segmentation in small datasets

Deep learning (DL) applied to breast tissue segmentation in magnetic resonance imaging (MRI) has received increased attention in the last decade, however, the domain shift which arises from different vendors, acquisition protocols, and biological heterogeneity, remains an important but challenging obstacle on the path towards clinical implementation. In this paper, we propose a novel Multi-level Semantic-guided Contrastive Domain Adaptation (MSCDA) framework to address this issue in an unsupervised manner. Our approach incorporates self-training with contrastive learning to align feature representations between domains. In particular, we extend the contrastive loss by incorporating pixel-to-pixel, pixel-to-centroid, and centroid-to-centroid contrasts to better exploit the underlying semantic information of the image at different levels. To resolve the data imbalance problem, we utilize a category-wise cross-domain sampling strategy to sample anchors from target images and build a hybrid memory bank to store samples from source images. We have validated MSCDA with a challenging task of cross-domain breast MRI segmentation between datasets of healthy volunteers and invasive breast cancer patients. Extensive experiments show that MSCDA effectively improves the model’s feature alignment capabilities between domains, outperforming state-of-the-art methods. Furthermore, the framework is shown to be label-efficient, achieving good performance with a smaller source dataset. The code is publicly available at https://github.com/ShengKuangCN/MSCDA.

PO-1641 Clinical evaluation of autosegmentation using AI with manual segmentation of breast tissue

PO-1641 Clinical evaluation of autosegmentation using AI with manual segmentation of breast tissue

Radiation dose estimates based on Monte Carlo simulation for spiral breast computed tomography imaging in a large cohort of patients.

Spiral breast computed tomography (BCT) equipped with a photon-counting detector (PCD) is a new radiological modality allowing for the compression-free acquisition of high-resolution 3-D datasets of the breast. Optimized dose exposu04170/re setups according to breast size were previously proposed but could not effectively be applied in a clinical environment due to ambiguity in measuring breast size. This study aims to report the standard radiation dose values in a large cohort of patients examined with BCT, and to provide a mathematical model to estimate radiation dose based on morphological features of the breast. This retrospective study was conducted on 1657 BCT examinations acquired between 2018 and 2021 from 829 participants (57±10 years, all female). Applying a dedicated breast tissue segmentation algorithm and Monte Carlo (MC) simulation, mean absorbed dose (MAD), mean glandular dose (MGD), mean skin dose (MSD), maximum glandular dose (maxGD), and maximum skin dose (maxSD) were calculated and related to morphological features such as breast volume, effective diameter, breast length, skin volume, and glandularity. Effective dose (ED) was calculated by applying the corresponding beam and tissue weighting factors, 1Sv/Gy and 0.12 per breast. Relevant morphological features predicting dose values were identified based on the Spearman's rank correlation coefficient. Exponential or bi-exponential models predicting the dose values as a function of morphological features were fitted by using a non-linear least squares (LS) method. The models were validated by assessing R2 and residual standard error (RSE). The most relevant morphological features for radiation dose estimation were the breast volume (correlation coefficient: -0.8), diameter (-0.7), and length (-0.6). The glandularity presented a weak-positive correlation (0.4) with MGD and maxGD due to the inhomogeneous distribution of the glandularity and absorbed dose in the 3-D breast volume. The standard MGDs were calculated to be 7.3±0.7, 6.5±0.3, and 5.9±0.3 mGy, MADs to 7.6±0.8, 6.8±0.3, and 6.2±0.3 mGy, maxSDs to 19.9±1.6, 19.5±0.5, and 18.9±0.5 mGy, and EDs to 0.88±0.08, 0.78±0.04, and 0.72±0.04 mSv for small, medium, and large breasts with average breast lengths of 5.9±1.6, 8.7±1.3, and 12.2±2.0cm, respectively. The estimated glandularity - 23.1±16.9, 12.5±11.4, and 6.9±7.3% from small to large breasts. The mathematical models were able to estimate the MAD, MGD, MSD, and maxSD as a function of each morphological feature with only upto 0.5 mGy RSE. We presented the typical morphological features and standard dose values according to the breast size acquired from a large patient cohort. We established radiation dose estimation models allowing accurate estimation of dose values including MGD with an acceptable RSE based on each of the easily measured morphological features of the breast. Clinicians could use the breast length to operate as a dosimetric alert of the scanner prior to a BCT scan. Radiation exposure for BCT was lower than diagnostic mammography (MG) and cone-beam breast CT (BCT).

Multiple-level thresholding for breast mass detection

Detection of breast mass plays a very important role in making the diagnosis of breast cancer. For faster detection of breast cancer caused by breast mass, we developed a novel and efficient patch-based breast mass detection system for mammography images. The proposed framework is comprised of three modules, including pre-processing, multiple-level breast tissue segmentation, and final breast mass detection. An improved Deeplabv3+ model for pectoral muscle removal is deployed in pre-processing. We then proposed a multiple-level thresholding segmentation method to segment breast mass and obtained the connected components (ConCs), where the corresponding image patch to each ConC is extracted for mass detection. In the final detection stage, each image patch is classified into breast mass and breast tissue background by trained deep learning models. The patches that are classified as breast mass are then taken as the candidates for breast mass. To reduce the false positive rate in the detection results, we applied the non-maximum suppression algorithm to combine the overlapped detection results. Once an image patch is considered a breast mass, the accurate detection result can then be retrieved from the corresponding ConC in the segmented images. Moreover, a coarse segmentation result can be simultaneously retrieved after detection. Compared to the state-of-the-art methods, the proposed method achieved comparable performance. On CBIS-DDSM, the proposed method achieved a detection sensitivity of 0.87 at 2.86 FPI (False Positive rate per Image), while the sensitivity reached 0.96 on INbreast with an FPI of only 1.29.

A deep learning approach for automatic detection, segmentation and classification of breast lesions from thermal images

PurposeThis study investigates implementation of deep learning (DL) approaches to breast tumor recognition based on thermal images. We propose to utilize Mask R-CNN technique on images by first assigning bounding boxes and then creating a border for each tumor volume to differentiate it from adjacent tissues and structures. In this manner, thermal images can be handled by a single DL model to successfully perform detection, classification, and segmentation of normal and abnormal breast tissues. MethodsThis study employs Mask R-CNN technique along with transfer learning models to accurately delineate breast volumes from adjacent tissues and structures. It is a novel study that uses a single DL model to carry out three steps of breast tumor diagnosis, namely detection, segmentation and classification of normal and abnormal tissues based on thermal images. ResultsTwo network architectures (ResNet-50 and ResNet-101) were trained for 60 epochs and were then evaluated based on their classification and segmentation performances. The testing process resulted in higher classification success for ResNet-50 backbone pretrained on COCO images (%97.1 accuracy). Detection and segmentation performances of this model were also higher with mAP of 0.921 and overlap score of 0.868. ConclusionsOur results indicate that the classification and segmentation performances of Mask R-CNN method on ResNet-50 architecture are better than the data reported in the literature for thermal breast image studies. It should be emphasized that our approach employs a single DL model that successfully performs detection, classification, and segmentation procedures for diagnosing normal and abnormal breast tissues.

Exploiting the Dixon Method for a Robust Breast and Fibro-Glandular Tissue Segmentation in Breast MRI.

Automatic breast and fibro-glandular tissue (FGT) segmentation in breast MRI allows for the efficient and accurate calculation of breast density. The U-Net architecture, either 2D or 3D, has already been shown to be effective at addressing the segmentation problem in breast MRI. However, the lack of publicly available datasets for this task has forced several authors to rely on internal datasets composed of either acquisitions without fat suppression (WOFS) or with fat suppression (FS), limiting the generalization of the approach. To solve this problem, we propose a data-centric approach, efficiently using the data available. By collecting a dataset of T1-weighted breast MRI acquisitions acquired with the use of the Dixon method, we train a network on both T1 WOFS and FS acquisitions while utilizing the same ground truth segmentation. Using the “plug-and-play” framework nnUNet, we achieve, on our internal test set, a Dice Similarity Coefficient (DSC) of 0.96 and 0.91 for WOFS breast and FGT segmentation and 0.95 and 0.86 for FS breast and FGT segmentation, respectively. On an external, publicly available dataset, a panel of breast radiologists rated the quality of our automatic segmentation with an average of 3.73 on a four-point scale, with an average percentage agreement of 67.5%.

Abstract 2736: Forecasting treatment response to neoadjuvant therapy in triple-negative breast cancer via an image-guided digital twin

Abstract Introduction: Patients with locally advanced, triple-negative breast cancer (TNBC) typically receive neoadjuvant therapy (NAT) to downstage the tumor and to improve the outcome of subsequent breast conservation surgery. There are currently no methods to accurately predict how a TNBC patient will respond to NAT before surgery. In this work, we applied a digital twin framework to address this unmet clinical need, by integrating quantitative magnetic resonance imaging (MRI) data with mechanism-based mathematical modeling. Methods: Multiparametric MRI was acquired in patients (N = 50) before, after 2 and 4 cycles of Adriamycin/Cyclophosphamide (A/C), and again after 12 cycles of Paclitaxel as part of the ARTEMIS (NCT02276433) trial. Within each imaging session, dynamic contrast-enhanced (DCE) MRI, diffusion-weighted imaging (DWI), and a pre-contrast T1-map were acquired. The images were processed by a pipeline consisting of motion correction, multiparametric image alignment, inter-visit image registration to align the tumor and surrounding breast tissue, tissue segmentation, and estimation of tumor cellularity from DWI. A mechanism-based mathematical model, a reaction-diffusion equation, is used to characterize the mobility of tumor cells via diffusion damped by mechanical tissue properties, tumor proliferation via logistic growth, and treatment-induced cell death via the delivery and decay of therapies. For each patient, pre-treatment images were used for model initialization. The model calibration and prediction were implemented with two strategies: 1) using images acquired during the A/C for calibration and predicting up to the end of A/C, and 2) using images acquired during and after the A/C for calibration and predicting up to the end of NAT. For strategy 1), we evaluated the model by comparing its predicted tumor volume and total tumor cellularity to the imaging measurements at the end of A/C. For strategy 2), we evaluated the model by comparing its predicted final response to the post-surgical pathological findings. Results: For strategy 1), our framework predicted the change of tumor volume and total tumor cellularity with Pearson correlation coefficients of 0.91 and 0.89, respectively. Regarding strategy 2), our framework achieved an area under the receiver operator characteristic curve of 0.88 for distinguishing pCR from non-pCR. As a comparison, imaging measurement of tumor volume at the end of A/C achieved an AUC of 0.79. Conclusion: Our approach successfully captures the patient-specific dynamics of TNBC response to NAT and provides an improved prediction of final response, which demonstrates the potential of a digital twin framework to be a powerful tool for predicting response to NAT. Once validated, the method will provide a unique opportunity for optimizing treatment plans on a patient-specific basis. Citation Format: Chengyue Wu, Angela M. Jarrett, Zijian Zhou, Nabil Elshafeey, Beatriz E. Adrada, Rosalind P. Candelaria, Rania Mohamed, Medine Boge, Lei Huo, Jason White, Debu Tripathy, Vicente Valero, Jennifer Litton, Clinton Yam, Jong Bum Son, Jingfei Ma, Gaiane M. Rauch, Thomas E. Yankeelov. Forecasting treatment response to neoadjuvant therapy in triple-negative breast cancer via an image-guided digital twin [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2736.

Deep Learning-Based Multi-Label Tissue Segmentation and Density Assessment from Mammograms

Objectives: Breast cancer is the most commonly diagnosed type of cancer among women and a common cause of cancer-related deaths. Early diagnosis and treatment of breast cancer is critical in disease prognosis. Breast density is known to have a correlation with breast cancer. In recent years, there has been an increasing interest in the investigation of computer-aided methods for early diagnosis of breast cancer. In this study, a new fully-automated deep learning-based cascaded model was proposed for breast density assessment. In the first stage, the segmentation of adipose, fibroglandular, and pectoral muscle tissues from the digitized film mammograms of the Digital Database for Screening Mammography (DDSM) was investigated using various types of U-nets. Features extracted from the breast tissue segmentation predictions were then used to assess breast density in the second stage. Material and methods: 66 and 296 mediolateral oblique mammograms were selected from DDSM dataset for segmentation and breast density assessment systems, respectively. Different U-nets with varying number of layers and filters were implemented and the model having the highest performance was determined. U-net performance was investigated using categorical cross-entropy, Dice, Tversky, Focal Tversky, and logarithmic cosine-hyperbolic Dice loss functions. The performances of U-nets having different types of connections were investigated. The performances of U-nets having pre-trained weights from VGG16, VGG19, and ResNet50 networks in the encoding path were also investigated. Segmentation results were improved by using an image processing pipeline based on morphological operators. Segmentation performance was presented in terms of accuracy, balanced accuracy, intersection over union, and Dice's similarity coefficient (DSC) metrics. The segmentation system predictions were then used to estimate mammographic density using a machine learning pipeline by extracting features related to the fibroglandular tissue percentage. Results: Using ResNet50-U-net on the test data, average DSC scores of 82.71%, 73.39%, and 95.30% were obtained for adipose, fibroglandular, and pectoral muscle tissue segmentation, respectively. The mammogram segmentation results are 3%-12% better than the current state-of-the-art DSC in the literature when considering all of the foreground tissues concurrently. A breast density classification accuracy of 76.01% was achieved on a separate mammogram dataset, which is comparable to the recent studies in the literature. Conclusion: The proposed system can be used for automatic segmentation of mammogram into adipose, fibroglandular, and pectoral muscle tissues. The segmentation model enables the estimation of the fibroglandular-adipose tissue interface, which is recently found to be an important region for breast cancer investigations. The proposed fully-automatic breast density assessment system has a comparable performance to the ones in the literature.

W-Net: Dense and diagnostic semantic segmentation of subcutaneous and breast tissue in ultrasound images by incorporating ultrasound RF waveform data.

We study the use of raw ultrasound waveforms, often referred to as the "Radio Frequency" (RF) data, for the semantic segmentation of ultrasound scans to carry out dense and diagnostic labeling. We present W-Net, a novel Convolution Neural Network (CNN) framework that employs the raw ultrasound waveforms in addition to the grey ultrasound image to semantically segment and label tissues for anatomical, pathological, or other diagnostic purposes. To the best of our knowledge, this is also the first deep-learning or CNN approach for segmentation that analyzes ultrasound raw RF data along with the grey image. We chose subcutaneous tissue (SubQ) segmentation as our initial clinical goal for dense segmentation since it has diverse intermixed tissues, is challenging to segment, and is an underrepresented research area. SubQ potential applications include plastic surgery, adipose stem-cell harvesting, lymphatic monitoring, and possibly detection/treatment of certain types of tumors. Unlike prior work, we seek to label every pixel in the image, without the use of a background class. A custom dataset consisting of hand-labeled images by an expert clinician and trainees are used for the experimentation, currently labeled into the following categories: skin, fat, fat fascia/stroma, muscle, and muscle fascia. We compared W-Net and attention variant of W-Net (AW-Net) with U-Net and Attention U-Net (AU-Net). Our novel W-Net's RF-Waveform encoding architecture outperformed regular U-Net and AU-Net, achieving the best mIoU accuracy (averaged across all tissue classes). We study the impact of RF data on dense labeling of the SubQ region, which is followed by the analyses of the generalization capability of the networks to patients and analysis on the SubQ tissue classes, determining that fascia tissues, especially muscle fascia in particular, are the most difficult anatomic class to recognize for both humans and AI algorithms. We present diagnostic semantic segmentation, which is semantic segmentation carried out for the purposes of direct diagnostic pixel labeling, and apply it to breast tumor detection task on a publicly available dataset to segment pixels into malignant tumor, benign tumor, and background tissue class. Using the segmented image we diagnose the patient by classifying the breast lesion as either benign or malignant. We demonstrate the diagnostic capability of RF data with the use of W-Net, which achieves the best segmentation scores across all classes.

Unsupervised domain adaptation for the segmentation of breast tissue in mammography images

Background and Objective: Breast density refers to the proportion of glandular and fatty tissue in the breast and is recognized as a useful factor assessing breast cancer risk. Moreover, the segmentation of the high-density glandular tissue from mammograms can assist medical professionals visualizing and localizing areas that may require additional attention. Developing robust methods to segment breast tissues is challenging due to the variations in mammographic acquisition systems and protocols. Deep learning methods are effective in medical image segmentation but they often require large quantities of labelled data. Unsupervised domain adaptation is an area of research that employs unlabelled data to improve model performance on variations of samples derived from different sources. Methods: First, a U-Net architecture was used to perform segmentation of the fatty and glandular tissues with labelled data from a single acquisition device. Then, adversarial-based unsupervised domain adaptation methods were used to incorporate single unlabelled target domains, consisting of images from a different machine, into the training. Finally, the domain adaptation model was extended to include multiple unlabelled target domains by combining a reconstruction task with adversarial training. Results: The adversarial training was found to improve the generalization of the initial model on new domain data, demonstrating clearly improved segmentation of the breast tissues. For training with multiple unlabelled domains, combining a reconstruction task with adversarial training improved the stability of the training and yielded adequate segmentation results across all domains with a single model. Conclusions: Results demonstrated the potential for adversarial-based domain adaptation with U-Net architectures for segmentation of breast tissue in mammograms coming from several devices and demonstrated that domain-adapted models could achieve a similar agreement with manual segmentations. It has also been found that combining adversarial and reconstruction-based methods can provide a simple and effective solution for training with multiple unlabelled target domains.

Development of U-Net Breast Density Segmentation Method for Fat-Sat MR Images Using Transfer Learning Based on Non-Fat-Sat Model

To develop a U-net deep learning method for breast tissue segmentation on fat-sat T1-weighted (T1W) MRI using transfer learning (TL) from a model developed for non-fat-sat images. The training dataset (N = 126) was imaged on a 1.5 T MR scanner, and the independent testing dataset (N = 40) was imaged on a 3 T scanner, both using fat-sat T1W pulse sequence. Pre-contrast images acquired in the dynamic-contrast-enhanced (DCE) MRI sequence were used for analysis. All patients had unilateral cancer, and the segmentation was performed using the contralateral normal breast. The ground truth of breast and fibroglandular tissue (FGT) segmentation was generated using a template-based segmentation method with a clustering algorithm. The deep learning segmentation was performed using U-net models trained with and without TL, by using initial values of trainable parameters taken from the previous model for non-fat-sat images. The ground truth of each case was used to evaluate the segmentation performance of the U-net models by calculating the dice similarity coefficient (DSC) and the overall accuracy based on all pixels. Pearson’s correlation was used to evaluate the correlation of breast volume and FGT volume between the U-net prediction output and the ground truth. In the training dataset, the evaluation was performed using tenfold cross-validation, and the mean DSC with and without TL was 0.97 vs. 0.95 for breast and 0.86 vs. 0.80 for FGT. When the final model developed with and without TL from the training dataset was applied to the testing dataset, the mean DSC was 0.89 vs. 0.83 for breast and 0.81 vs. 0.81 for FGT, respectively. Application of TL not only improved the DSC, but also decreased the required training case number. Lastly, there was a high correlation (R2 > 0.90) for both the training and testing datasets between the U-net prediction output and ground truth for breast volume and FGT volume. U-net can be applied to perform breast tissue segmentation on fat-sat images, and TL is an efficient strategy to develop a specific model for each different dataset.

Segmentation of whole breast and fibroglandular tissue using nnU-Net in dynamic contrast enhanced MR images

PurposeSegmentation of the whole breast and fibroglandular tissue (FGT) is important for quantitatively analyzing the breast cancer risk in the dynamic contrast-enhanced magnetic resonance (DCE-MR) images. The purpose of this study is to improve the accuracy and efficiency of the segmentation of the whole breast and FGT in 3-D fat-suppressed DCE-MR images with a versatile deep learning (DL) framework. MethodsWe randomly collected 100 breast DCE-MR scans from Shanghai Cancer Hospital of Fudan University. The MR scans in the dataset were different in both the spatial resolution and the MR scanners employed. Furthermore, four breast density categories were assessed by radiologists based on Breast Imaging Reporting and Data System (BI-RADS) of American College of Radiology. The dataset was separated into the training and the testing sets, while keeping a balanced distribution of scans with different imaging parameters and density categories. The nnU-Net has been recently proposed to automatically adapt preprocessing strategies and network architectures for a given medical image dataset, thus showing a great potential in the systematic adaptation of DL methods to different datasets. In this study, we applied the nnU-Net to segment the whole breast and FGT in 3-D fat-suppressed DCE-MR images. Five-fold cross validation was employed to train and validate the segmentation method. ResultsThe segmentation performance was evaluated with the volume and surface agreement metrics between the DL-based automatic and the manually delineated masks, as quantified with the following measures: the average Dice volume overlap (0.968 ± 0.017 and 0.877 ± 0.081), the average surface distances (0.201 ± 0.080 mm and 0.310 ± 0.043 mm), and the Pearson correlation coefficient of masks (0.995 and 0.972) between the automatic and the manually delineated masks, as calculated for the whole breast and the FGT segmentation, respectively. The correlation coefficient between the breast densities obtained with the DL-based segmentation and the manual delineation was 0.981. There was a positive bias of 0.8% (DL-based relative to manual) in breast density measurement with the Bland-Altman plot. The execution time of the DL-based segmentation was approximately 20 s for the whole breast segmentation and 15 s for the FGT segmentation. ConclusionsOur DL-based segmentation framework using nnU-Net could robustly achieve high accuracy and efficiency across variable MR imaging settings without extra pre- or post-processing procedures. It would be useful for developing DCE-MR-based CAD systems to quantify breast cancer risk and to be integrated into the clinical workflow.

Automatic breast tissue segmentation in MRIs with morphology snake and deep denoiser training via extended Stein's unbiased risk estimator.

Accurate segmentation of the breast tissue is a significant challenge in the analysis of breast MR images, especially analysis of breast images with low contrast. Most of the existing methods for breast segmentation are semi-automatic and limited in their ability to achieve accurate results. This is because of difficulties in removing landmarks from noisy magnetic resonance images (MRI). Especially, when tumour is imaged for scanning, how to isolate the tumour region from chest will directly affect the accuracy for tumour to be detected. Due to low intensity levels and the close connection between breast and chest portion in MRIs, this study proposes an innovative, fully automatic and fast segmentation approach which combines histogram with inverse Gaussian gradient for morphology snakes, along with extended Stein's unbiased risk estimator (eSURE) applied for unsupervised learning of deep neural network Gaussian denoisers, aimed at accurate identification of landmarks such as chest and breast.

Fully Automated Breast Density Segmentation and Classification Using Deep Learning.

Breast density estimation with visual evaluation is still challenging due to low contrast and significant fluctuations in the mammograms’ fatty tissue background. The primary key to breast density classification is to detect the dense tissues in the mammographic images correctly. Many methods have been proposed for breast density estimation; nevertheless, most of them are not fully automated. Besides, they have been badly affected by low signal-to-noise ratio and variability of density in appearance and texture. This study intends to develop a fully automated and digitalized breast tissue segmentation and classification using advanced deep learning techniques. The conditional Generative Adversarial Networks (cGAN) network is applied to segment the dense tissues in mammograms. To have a complete system for breast density classification, we propose a Convolutional Neural Network (CNN) to classify mammograms based on the standardization of Breast Imaging-Reporting and Data System (BI-RADS). The classification network is fed by the segmented masks of dense tissues generated by the cGAN network. For screening mammography, 410 images of 115 patients from the INbreast dataset were used. The proposed framework can segment the dense regions with an accuracy, Dice coefficient, Jaccard index of 98%, 88%, and 78%, respectively. Furthermore, we obtained precision, sensitivity, and specificity of 97.85%, 97.85%, and 99.28%, respectively, for breast density classification. This study’s findings are promising and show that the proposed deep learning-based techniques can produce a clinically useful computer-aided tool for breast density analysis by digital mammography.