scholarly journals Independent validation of machine learning in diagnosing breast Cancer on magnetic resonance imaging within a single institution

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Yu Ji ◽  
Hui Li ◽  
Alexandra V. Edwards ◽  
John Papaioannou ◽  
Wenjuan Ma ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Magnetic Resonance Imaging ◽  
Machine Learning ◽  
Magnetic Resonance ◽  
Breast Lesions ◽  
Resonance Imaging ◽  
Single Institution ◽  
Dce Mri ◽  
Benign Breast Lesions ◽  
Mass Lesions

Abstract Background As artificial intelligence methods for the diagnosis of disease advance, we aimed to evaluate machine learning in the predictive task of distinguishing between malignant and benign breast lesions on an independent clinical magnetic resonance imaging (MRI) dataset within a single institution for subsequent use as a computer aid for radiologists. Methods Computer analysis was conducted on consecutive dynamic contrast-enhanced MRI (DCE-MRI) studies from 1483 breast cancer and 496 benign patients who underwent MRI examinations between February 2015 and October 2017; with the age ranges of the cancer and benign patients being 19 to 77 and 16 to 76 years old, respectively. Cases were separated into a training dataset (years 2015 & 2016; 1444 cases) and an independent testing dataset (year 2017; 535 cases) based solely on MRI examination date. After radiologist indication of the lesion, the computer automatically segmented and extracted radiomic features, which were subsequently merged with a support-vector machine (SVM) to yield a lesion signature. Area under the receiving operating characteristic (ROC) curve (AUC) with 95% confidence intervals (CI) served as the primary figure of merit in the statistical evaluation for this clinical classification task. Results In the task of distinguishing malignant and benign breast lesions DCE-MRI, the trained predictive model yielded an AUC value of 0.89 (95% CI: 0.858, 0.922) on the independent image set. AUC values of 0.88 (95% CI: 0.845, 0.926) and 0.90 (95% CI: 0.837, 0.940) were obtained for mass lesions only and non-mass lesions only, respectively. Compared with actual clinical management decisions, the predictive model achieved 99.5% sensitivity with 9.6% fewer recommended biopsies. Conclusion On an independent, consecutive clinical dataset within a single institution, a trained machine learning system yielded promising performance in distinguishing between malignant and benign breast lesions.

Washout of Mass-Like Benign Breast Lesions at Dynamic Magnetic Resonance Imaging

2012 ◽  
Vol 36 (3) ◽  
pp. 301-305 ◽  
Author(s):  
Penampai Tannaphai ◽  
Rubina Manuela Trimboli ◽  
Luca Alessandro Carbonaro ◽  
Sara Viganò ◽  
Giovanni Di Leo ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Dynamic Magnetic Resonance Imaging ◽  
Breast Lesions ◽  
Resonance Imaging ◽  
Dynamic Magnetic Resonance ◽  
Benign Breast Lesions

scholarly journals Machine learning magnetic resonance imaging radiomics predicts axillary lymph node metastasis in invasive breast cancer

2020 ◽  
Author(s):  
Yunfang Yu ◽  
Zifan He ◽  
Jie Ouyang ◽  
Yujie Tan ◽  
Yong-Jian Chen ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Magnetic Resonance Imaging ◽  
Machine Learning ◽  
Lymph Node ◽  
Clinical Practice ◽  
Magnetic Resonance ◽  
Invasive Breast Cancer ◽  
Axillary Lymph Node ◽  
Axillary Lymph ◽  
Resonance Imaging

Abstract In current clinical practice, the standard evaluation for axillary lymph node (ALN) status in breast cancer is based on the invasive procedure and many patients will suffer from operative associated complications. Hence, a novel signature incorporated tumor and lymph node magnetic resonance imaging (MRI) radiomics, clinical and pathological characteristics, and molecular subtypes based on the machine learning approach was established to accurately identify ALN metastasis in early-stage invasive breast cancer patients. Although the misjudgment of ALN status by clinicians according to preoperative MRI are common during clinical practice and even the senior radiologists make mistakes sometimes, this multiomic radiomic signature showed the superiority over clinicians and could precisely discriminate ALN metastasis among different molecular subtype patients. Furthermore, the association between MRI radiomic features and tumor-microenvironment features including immune cells, long non-coding RNAs, and types of methylated sites were found, which revealed the potential biological underpinning of MRI radiomics.

scholarly journals Evaluation of diagnostic value of ultrasound and magnetic resonance imaging for malignant and benign breast lesions

2018 ◽  
Vol 2 (5) ◽  
Author(s):  
Yanni Zeng ◽  
Hongwei Zhang ◽  
Jiuxia Zhang ◽  
Yan Yu ◽  
Liangjin Liu
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Statistical Significance ◽  
Diagnostic Value ◽  
Breast Lesions ◽  
Resonance Imaging ◽  
Benign Lesions ◽  
Specificity And Sensitivity ◽  
Treatment Data ◽  
Benign Breast Lesions

[Abstract] Objectives: To investigate diagnostic value of ultrasound and magnetic resonance imaging (MRI) for malignant and benign breast lesions. Methods: Retrospective analysis of treatment data of 48 patients diagnosed with malignant and benign breast lesions in our hospital, collected from December 2017 to November 2018. A total number of 56 breast masses were examined by both ultrasound and MRI, and were compared with postoperative pathological biopsy results. Results: Postoperative pathological biopsy results showed that there were 26 and 30 malignant and benign lesions respectively. Comparison of MRI curve type of malignant and benign lesions showed statistical significance (P<0.05). By comparison with pathological biopsy results, specificity and sensitivity of ultrasound diagnosis were 83.33% (25/30) and 84.61% (22/26) respectively; specificity and sensitivity of MRI diagnosis were 96.66% (29/30) and 92.30% (24/26) respectively. Conclusions: Ultrasonographic examination of malignant and benign breast lesions is straight-forward, simple and inexpensive. Accuracy, specificity and sensitivity of MRI are significantly higher than ultrasound in examining malignant and benign breast lesions, this can reduce misdiagnosis.

scholarly journals Differentiation of breast tuberculosis and breast cancer using diffusion-weighted, T2-weighted and dynamic contrast-enhanced magnetic resonance imaging

2018 ◽  
Vol 22 (2) ◽  
Author(s):  
Dibuseng P. Ramaema ◽  
Richard J. Hift
Keyword(s):  
Breast Cancer ◽  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Resonance Imaging ◽  
Adc Value ◽  
Dce Mri ◽  
Dynamic Contrast Enhanced ◽  
Diffusion Weighted ◽  
Breast Tuberculosis ◽  
Contrast Enhanced

Background: The use of multi-parametric magnetic resonance imaging (MRI) in the evaluation of breast tuberculosis (BTB).Objectives: To evaluate the value of diffusion-weighted imaging (DWI), T2-weighted (T2W) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiating breast cancer (BCA) from BTB.Method: We retrospectively studied images of 17 patients with BCA who had undergone preoperative MRI and 6 patients with pathologically proven BTB who underwent DCE-MRI during January 2014 to January 2015.Results: All patients were female, with the age range of BTB patients being 23–43 years and the BCA patients being 31–74 years. Breast cancer patients had a statistically significant lower mean apparent diffusion coefficient (ADC) value (1072.10 +/- 365.14), compared to the BTB group (1690.77 +/- 624.05, p = 0.006). The mean T2-weighted signal intensity (T2SI) was lower for the BCA group (521.56 +/- 233.73) than the BTB group (787.74 +/- 196.04, p = 0.020). An ADC mean cut-off value of 1558.79 yielded 66% sensitivity and 94% specificity, whilst the T2SI cut-off value of 790.20 yielded 83% sensitivity and 83% specificity for differentiating between BTB and BCA. The homogeneous internal enhancement for focal mass was seen in BCA patients only.Conclusion: Multi-parametric MRI incorporating the DWI, T2W and DCE-MRI may be a useful tool to differentiate BCA from BTB.

scholarly journals Non-Invasive Assessment of Breast Cancer Molecular Subtypes with Multiparametric Magnetic Resonance Imaging Radiomics

2020 ◽  
Vol 9 (6) ◽  
pp. 1853
Author(s):  
Doris Leithner ◽  
Marius E. Mayerhoefer ◽  
Danny F. Martinez ◽  
Maxine S. Jochelson ◽  
Elizabeth A. Morris ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Molecular Subtypes ◽  
Resonance Imaging ◽  
Luminal A ◽  
Linear Discriminant ◽  
Dce Mri ◽  
Non Invasive ◽  
Multiparametric Magnetic Resonance Imaging

We evaluated the performance of radiomics and artificial intelligence (AI) from multiparametric magnetic resonance imaging (MRI) for the assessment of breast cancer molecular subtypes. Ninety-one breast cancer patients who underwent 3T dynamic contrast-enhanced (DCE) MRI and diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) mapping were included retrospectively. Radiomic features were extracted from manually drawn regions of interest (n = 704 features per lesion) on initial DCE-MRI and ADC maps. The ten best features for subtype separation were selected using probability of error and average correlation coefficients. For pairwise comparisons with >20 patients in each group, a multi-layer perceptron feed-forward artificial neural network (MLP-ANN) was used (70% of cases for training, 30%, for validation, five times each). For all other separations, linear discriminant analysis (LDA) and leave-one-out cross-validation were applied. Histopathology served as the reference standard. MLP-ANN yielded an overall median area under the receiver-operating-characteristic curve (AUC) of 0.86 (0.77–0.92) for the separation of triple negative (TN) from other cancers. The separation of luminal A and TN cancers yielded an overall median AUC of 0.8 (0.75–0.83). Radiomics and AI from multiparametric MRI may aid in the non-invasive differentiation of TN and luminal A breast cancers from other subtypes.

Sign in / Sign up

Export Citation Format

Share Document