scholarly journals MRI Breast Tumor Segmentation Using Different Encoder and Decoder CNN Architectures

Computers ◽  
2019 ◽  
Vol 8 (3) ◽  
pp. 52 ◽  
Author(s):  
Adoui ◽  
Mahmoudi ◽  
Larhmam ◽  
Benjelloun
Keyword(s):  
Breast Cancer ◽  
Breast Tumor ◽  
High Performance ◽  
Tumor Segmentation ◽  
Learning Approaches ◽  
Graphic Processing Units ◽  
Training Time ◽  
Dce Mri ◽  
Experienced Radiologist ◽  
Before And After

Breast tumor segmentation in medical images is a decisive step for diagnosis and treatment follow-up. Automating this challenging task helps radiologists to reduce the high manual workload of breast cancer analysis. In this paper, we propose two deep learning approaches to automate the breast tumor segmentation in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) by building two fully convolutional neural networks (CNN) based on SegNet and U-Net. The obtained models can handle both detection and segmentation on each single DCE-MRI slice. In this study, we used a dataset of 86 DCE-MRIs, acquired before and after two cycles of chemotherapy, of 43 patients with local advanced breast cancer, a total of 5452 slices were used to train and validate the proposed models. The data were annotated manually by an experienced radiologist. To reduce the training time, a high-performance architecture composed of graphic processing units was used. The model was trained and validated, respectively, on 85% and 15% of the data. A mean intersection over union (IoU) of 68.88 was achieved using SegNet and 76.14% using U-Net architecture.

scholarly journals Semi-automatic segmentation from intrinsically-registered 18F-FDG–PET/MRI for treatment response assessment in a breast cancer cohort: comparison to manual DCE–MRI

2019 ◽  
Vol 33 (2) ◽  
pp. 317-328
Author(s):  
Maren Marie Sjaastad Andreassen ◽  
Pål Erik Goa ◽  
Torill Eidhammer Sjøbakk ◽  
Roja Hedayati ◽  
Hans Petter Eikesdal ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Treatment Response ◽  
Neoadjuvant Treatment ◽  
Gaussian Mixture ◽  
Tumor Segmentation ◽  
Dice Similarity Coefficient ◽  
Adc Value ◽  
Dce Mri ◽  
18F Fdg ◽  
Tumor Area

Abstract Objectives To investigate the reliability of simultaneous positron emission tomography and magnetic resonance imaging (PET/MRI)-derived biomarkers using semi-automated Gaussian mixture model (GMM) segmentation on PET images, against conventional manual tumor segmentation on dynamic contrast-enhanced (DCE) images. Materials and methods Twenty-four breast cancer patients underwent PET/MRI (following 18F-fluorodeoxyglucose (18F-FDG) injection) at baseline and during neoadjuvant treatment, yielding 53 data sets (24 untreated, 29 treated). Two-dimensional tumor segmentation was performed manually on DCE–MRI images (manual DCE) and using GMM with corresponding PET images (GMM–PET). Tumor area and mean apparent diffusion coefficient (ADC) derived from both segmentation methods were compared, and spatial overlap between the segmentations was assessed with Dice similarity coefficient and center-of-gravity displacement. Results No significant differences were observed between mean ADC and tumor area derived from manual DCE segmentation and GMM–PET. There were strong positive correlations for tumor area and ADC derived from manual DCE and GMM–PET for untreated and treated lesions. The mean Dice score for GMM–PET was 0.770 and 0.649 for untreated and treated lesions, respectively. Discussion Using PET/MRI, tumor area and mean ADC value estimated with a GMM–PET can replicate manual DCE tumor definition from MRI for monitoring neoadjuvant treatment response in breast cancer.

Effect of tibolone on tumor angiogenesis parameters in short-term treated women

2006 ◽  
Vol 24 (18_suppl) ◽  
pp. 10519-10519
Author(s):  
F. E. Frankenne ◽  
C. Munaut ◽  
F. Wijzen ◽  
J. Planellas Gomez ◽  
L. Kloosterboer ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Breast Tumor ◽  
Vascular Density ◽  
Potential Mechanism ◽  
Short Term ◽  
Related Factors ◽  
Increased Risk ◽  
Before And After ◽  
Pai 1 ◽  
Neutral Effect

10519 Background: Large recent studies have shown increased mammographic density, breast pain and a marginally increased risk of breast cancer in women treated with continuous combined estrogen/progestogen (ccEPT), while tibolone has a minimal effect. A placebo-controled clinical trial (STEM), approved by several ethical comities, has been initiated to assess effect and potential mechanism of action of tibolone on breast tumor. We present the results on 12 angiogenesis-related parameters. Methods: 102 women with breast cancer who had to undergo a mastectomy were treated for 2 weeks with tibolone 2.5 mg/day or placebo. RNA was extracted from tumor snap-frozen biopsies (paired before and after treatments) and reverse-transcripted into 2 cDNA batches on which 6 distinct PCRs were performed in duplicate per parameter/sample. A vascular density (VD) index was determined on CD31 specifically stained tumor sections by counting 10 fields per section on 5 distinct sections per biopsy. Results: The validity of the VD evaluation is highlighted by the high correlation coefficient (r = 0.81) between the data within the placebo group at a 2 week interval. In the tumor samples, no change of the VD index was seen. Since it could be argued that VD index could not correctly reflect a modulation of angiogenesis after only 14 days of treatment, we looked at a potential change of the expression of a large array of angiogenesis-related factors (VEGFs, VEGFRs, PAI-1). No modification of the expression in any of these factors was observed. Conclusions: Our data lead to conclude that, in this study, tibolone displays a neutral effect on angiogenesis in breast tumor. [Table: see text]

scholarly journals Review methods for breast cancer detection using artificial intelligence and deep learning methods

2021 ◽  
pp. 98-102
Author(s):  
Maryam Naderan
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Cancer Detection ◽  
High Performance ◽  
False Negative ◽  
Main Idea ◽  
New Approach ◽  
Training Time ◽  
Proposed Model ◽  
Convolutional Autoencoder

Nowadays, there are many related works and methods that use Neural Networks to detect the breast cancer. However, usually they do not take into account the training time and the result of False Negative (FN) while training the model. The main idea of this paper is to compare already existing methods for detecting the breast cancer using Deep Learning Algorithms. Moreover, since the breast cancer is one of the most common lethal cancers and early detection helps prevent complications, we propose a new approach and the use of the convolutional autoencoder. This proposed model has shown high performance with sensitivity, precision, and accuracy of 93,50%, 91,60% and 93% respectively.

scholarly journals Radiomics of Tumor Heterogeneity in Longitudinal Dynamic Contrast-Enhanced Magnetic Resonance Imaging for Predicting Response to Neoadjuvant Chemotherapy in Breast Cancer

2021 ◽  
Vol 8 ◽  
Author(s):  
Ming Fan ◽  
Hang Chen ◽  
Chao You ◽  
Li Liu ◽  
Yajia Gu ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Predictive Models ◽  
Breast Tumor ◽  
Tumor Heterogeneity ◽  
Molecular Subtype ◽  
Resonance Imaging ◽  
Dce Mri ◽  
Longitudinal Dynamic ◽  
Contrast Enhanced

Breast tumor morphological and vascular characteristics can be changed during neoadjuvant chemotherapy (NACT). The early changes in tumor heterogeneity can be quantitatively modeled by longitudinal dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), which is useful in predicting responses to NACT in breast cancer. In this retrospective analysis, 114 female patients with unilateral unifocal primary breast cancer who received NACT were included in a development (n = 61) dataset and a testing dataset (n = 53). DCE-MRI was performed for each patient before and after treatment (two cycles of NACT) to generate baseline and early follow-up images, respectively. Feature-level changes (delta) of the entire tumor were evaluated by calculating the relative net feature change (deltaRAD) between baseline and follow-up images. The voxel-level change inside the tumor was evaluated, which yielded a Jacobian map by registering the follow-up image to the baseline image. Clinical information and the radiomic features were fused to enhance the predictive performance. The area under the curve (AUC) values were assessed to evaluate the prediction performance. Predictive models using radiomics based on pre- and post-treatment images, Jacobian maps and deltaRAD showed AUC values of 0.568, 0.767, 0.630 and 0.726, respectively. When features from these images were fused, the predictive model generated an AUC value of 0.771. After adding the molecular subtype information in the fused model, the performance was increased to an AUC of 0.809 (sensitivity of 0.826 and specificity of 0.800), which is significantly higher than that of the baseline imaging- and Jacobian map-based predictive models (p = 0.028 and 0.019, respectively). The level of tumor heterogeneity reduction (evaluated by texture feature) is higher in the NACT responders than in the nonresponders. The results suggested that changes in DCE-MRI features that reflect a reduction in tumor heterogeneity following NACT could provide early prediction of breast tumor response. The prediction was improved when the molecular subtype information was combined into the model.

Automated breast cancer lesion detection on breast MRI using artificial intelligence.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. e14612-e14612 ◽  
Author(s):  
Nataly Tapia Negrete ◽  
Ruquaiyah Takhtawala ◽  
Madeleine Shaver ◽  
Turkay Kart ◽  
Yang Zhang ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Artificial Intelligence ◽  
Breast Tumor ◽  
Breast Mri ◽  
Lesion Detection ◽  
Tumor Segmentation ◽  
Dice Similarity Coefficient ◽  
Data Set ◽  
Post Contrast

e14612 Background: Over 40,000 women in the US will die from breast cancer. Early detection of cancer is crucial and is a potential avenue to improve survival. The objective of this research study is to develop a convoluted neural network (CNN), a subset of artificial intelligence, in order to enhance computerized detection of breast lesions on MRIs. Methods: This is an institutional review board approved retrospective study with post contrast MRI data from 238 patients. Breast tumor segmentation was automated with a hybrid 3D/2D CNN designed adapted from U-net, a popular neural network architecture in biomedical image analysis. T1 post-contrast MRI volumes were used to train the network. The data set was separated into training (80%) and validation (20%) sets. Re-sampling and normalization using z-scores were applied to each volume before training. Contracting and expanding arms of the model consist of successive convolutions followed by batch normalization and ReLU operations. Ground truth was established through manual segmentation and previously conducted readings of the images used to train our network. Results: A 5-fold cross validation was performed for analysis. The Dice similarity coefficient was used to assess segmentation accuracy. The hybrid 3D/2D U-Net architecture yielded a Dice score of 0.753 and a Pearson correlation of 0.548 for the breast tumor segmentation. Conclusions: These results demonstrated the feasibility for artificial intelligence applications in accurately identifying the presence of lesions on breast MRI images.

scholarly journals Assessing PD-L1 Expression Status Using Radiomic Features from Contrast-Enhanced Breast MRI in Breast Cancer Patients: Initial Results

Cancers ◽  
2021 ◽  
Vol 13 (24) ◽  
pp. 6273
Author(s):  
Roberto Lo Gullo ◽  
Hannah Wen ◽  
Jeffrey S. Reiner ◽  
Raza Hoda ◽  
Varadan Sevilimedu ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Breast Mri ◽  
Standard Of Care ◽  
Tumor Segmentation ◽  
Imaging Features ◽  
Dce Mri ◽  
Contrast Enhanced ◽  
Expression Status

The purpose of this retrospective study was to assess whether radiomics analysis coupled with machine learning (ML) based on standard-of-care dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict PD-L1 expression status in patients with triple negative breast cancer, and to compare the performance of this approach with radiologist review. Patients with biopsy-proven triple negative breast cancer who underwent pre-treatment breast MRI and whose PD-L1 status was available were included. Following 3D tumor segmentation and extraction of radiomic features, radiomic features with significant differences between PD-L1+ and PD-L1− patients were determined, and a final predictive model to predict PD-L1 status was developed using a coarse decision tree and five-fold cross-validation. Separately, all lesions were qualitatively assessed by two radiologists independently according to the BI-RADS lexicon. Of 62 women (mean age 47, range 31–81), 27 had PD-L1− tumors and 35 had PD-L1+ tumors. The final radiomics model to predict PD-L1 status utilized three MRI parameters, i.e., variance (FO), run length variance (RLM), and large zone low grey level emphasis (LZLGLE), for a sensitivity of 90.7%, specificity of 85.1%, and diagnostic accuracy of 88.2%. There were no significant associations between qualitative assessed DCE-MRI imaging features and PD-L1 status. Thus, radiomics analysis coupled with ML based on standard-of-care DCE-MRI is a promising approach to derive prognostic and predictive information and to select patients who could benefit from anti-PD-1/PD-L1 treatment.

Sign in / Sign up

Export Citation Format

Share Document