Computer-Aided Diagnosis Evaluation of the Correlation Between Magnetic Resonance Imaging With Molecular Subtypes in Breast Cancer

Meng, Wei; Sun, Yihong; Qian, Haibin; Chen, Xiaodan; Yu, Qiujie; Nanding, Abiyasi; Yan, Shaolei; Peng, Haiyong; Zhang, Hongxia; Zhang, Xiu-Shi

doi:10.3389/fonc.2021.693339

Cited by 7 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most studies focused on predicting the Ki-67 status (an index that has important implications for molecular subtyping and treatment of breast cancer patients), 175,193,199 and identifying the immunohistochemical subtypes (Luminal A, Lumi-nal B, HER2-positive, and triple negative) of breast cancer. 178,179,186,191,192 Both radiomics features and DL features have been experimented with, and promising performance has been achieved.…”

Section: Breast Cancer Diagnosismentioning

confidence: 99%

Artificial intelligence in multiparametric magnetic resonance imaging: A review

Liu

et al. 2022

Medical Physics

View full text Add to dashboard Cite

Multiparametric magnetic resonance imaging (mpMRI) is an indispensable tool in the clinical workflow for the diagnosis and treatment planning of various diseases. Machine learning-based artificial intelligence (AI) methods, especially those adopting the deep learning technique, have been extensively employed to perform mpMRI image classification, segmentation, registration, detection, reconstruction, and super-resolution. The current availabilities of increasing computational power and fast-improving AI algorithms have empowered numerous computer-based systems for applying mpMRI to disease diagnosis, imaging-guided radiotherapy, patient risk and overall survival time prediction, and the development of advanced quantitative imaging technology for magnetic resonance fingerprinting.However,the wide application of these developed systems in the clinic is still limited by a number of factors, including robustness, reliability, and interpretability. This survey aims to provide an overview for new researchers in the field as well as radiologists with the hope that they can understand the general concepts, main application scenarios, and remaining challenges of AI in mpMRI.

show abstract

Section: Breast Cancer Diagnosismentioning

confidence: 99%

Artificial intelligence in multiparametric magnetic resonance imaging: A review

Liu

et al. 2022

Medical Physics

View full text Add to dashboard Cite

show abstract

“…46 Additionally, they are adding more credence to the existing reports within various breast MRI radiomics paradigms. [17][18][19][20] Regarding breast MRI diagnosis, in a previously proposed CADx based on ensemble methods for feature selection and classification, AdaBoost has achieved a high performance (AUC = 0.96) in differentiating malignant and benign lesions by means of DCE MRI radiomic features. 21 Beyond the exploitation of ensemble learning methods, the authors have also made use of wavelet features.…”

Section: Discussionmentioning

confidence: 99%

“… 15 These strategies have proven very useful in modelling heterogeneous datasets of any size and complexity, 7 while also excel at trading off the approximation and estimation errors compared to the more conventional ML approaches. 16 Particularly, Boosting Ensemble Classifiers have shown to outperform other classification techniques within breast mpMRI radiomics, for molecular subtypes recognition, 17 , 18 prediction of sentinel lymph node metastasis, 19 and early prediction of treatment response and survival outcomes. 20 Additionally, their predictive efficiency for differentiating benign from malignant breast lesions has shown promise within DCE MRI radiomics alone in a recent study (AUC = 0.96), 21 but this have not yet been evaluated within mpMRI datatsets.…”

Section: Introductionmentioning

confidence: 99%

Breast Cancer Classification on Multiparametric MRI – Increased Performance of Boosting Ensemble Methods

Vamvakas

Logothetis

Vassiou

et al. 2022

Technol Cancer Res Treat

View full text Add to dashboard Cite

Introduction: This study aims to assess the utility of Boosting ensemble classification methods for increasing the diagnostic performance of multiparametric Magnetic Resonance Imaging (mpMRI) radiomic models, in differentiating benign and malignant breast lesions. Methods: The dataset includes mpMR images of 140 female patients with mass-like breast lesions (70 benign and 70 malignant), consisting of Dynamic Contrast Enhanced (DCE) and T2-weighted sequences, and the Apparent Diffusion Coefficient (ADC) calculated from the Diffusion Weighted Imaging (DWI) sequence. Tumor masks were manually defined in all consecutive slices of the respective MRI volumes and 3D radiomic features were extracted with the Pyradiomics package. Feature dimensionality reduction was based on statistical tests and the Boruta wrapper. Hierarchical Clustering on Spearman's rank correlation coefficients between features and Random Forest classification for obtaining feature importance, were implemented for selecting the final feature subset. Adaptive Boosting (AdaBoost), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) classifiers, were trained and tested with bootstrap validation in differentiating breast lesions. A Support Vector Machine (SVM) classifier was also exploited for comparison. The Receiver Operator Characteristic (ROC) curves and DeLong's test were utilized to evaluate the classification performances. Results: The final feature subset consisted of 5 features derived from the lesion shape and the first order histogram of DCE and ADC images volumes. XGboost and LGBM achieved statistically significantly higher average classification performances [AUC = 0.95 and 0.94 respectively], followed by Adaboost [AUC = 0.90], GB [AUC = 0.89] and SVM [AUC = 0.88]. Conclusion: Overall, the integration of Ensemble Learning methods within mpMRI radiomic analysis can improve the performance of computer-assisted diagnosis of breast cancer lesions.

show abstract

“…Furthermore, the potential correlations between feature categories and clinical outcomes must be carefully considered. For instance, texture analysis has been correlated with gene expression pro les and can serve as a non-invasive surrogate for the tumor's molecular subtype [13]. Similarly, rst-order features have been linked with therapeutic response, offering a non-invasive means of monitoring treatment e cacy [14].…”

Section: Interpretation Of Radiomic Feature Extraction and Classi Cationmentioning

confidence: 99%

Revolutionizing Treatment Planning: Habitat-Based Radiomics for Lateral Lymph Node Metastasis Prediction

刘,

叮,

董

et al. 2024

Preprint

View full text Add to dashboard Cite

Background:Recent advancements in tumor microenvironment analysis have significantly impacted immunotherapy strategies, particularly in thyroid papillary carcinoma. This study focuses on the value of habitat-based radiomics for predicting lateral lymph node metastasis, a crucial factor in treatment planning and prognosis. Methods:The study selected participants with thyroid papillary carcinoma undergoing their first surgical treatment. Criteria included complete clinical data and enhanced CT imaging. Medical images were normalized and resampled for fixed-resolution pixel values. Radiomics features, classified into geometry, intensity, and texture, were extracted using the pyradiomics tool. Feature selection involved Intraclass Correlation Coefficient (ICC) and LASSO regression. Machine learning models such as Support Vector Machine (SVM), RandomForest (RF), and ExtraTrees (ET) were used to construct radiomic and habitat signatures with a specific focus on identifying lateral lymph node metastasis. Results:The habitat-based models demonstrated high efficacy in predicting lateral lymph node metastasis. The Habitat Signature showed higher accuracy (94.6% for SVM, 94.6% for RF, 91.9% for ET) and Area Under the Curve (AUC) values (0.988 for SVM, 0.961 for RF, 0.982 for ET) compared to the Radiomics Signature, specifically in identifying metastatic nodes. The Habitat model also had superior calibration performance, as evidenced by Hosmer-Lemeshow test statistics in training, validation, and test cohorts. Decision curve analysis indicated the Habitat Signature's potential for significant clinical benefit in predicting lateral lymph node involvement. Conclusion:Habitat-based radiomics analysis provides an accurate and efficient approach for predicting lateral lymph node metastasis in thyroid papillary carcinoma. This method enhances the predictive accuracy, facilitating better personalized treatment strategies in immunotherapy settings. It offers a promising tool in personalized medicine, especially for planning targeted treatment and assessing prognosis in thyroid cancer patients.

show abstract

Computer-Aided Diagnosis Evaluation of the Correlation Between Magnetic Resonance Imaging With Molecular Subtypes in Breast Cancer

Cited by 7 publications

References 27 publications

Artificial intelligence in multiparametric magnetic resonance imaging: A review

Artificial intelligence in multiparametric magnetic resonance imaging: A review

Breast Cancer Classification on Multiparametric MRI – Increased Performance of Boosting Ensemble Methods

Revolutionizing Treatment Planning: Habitat-Based Radiomics for Lateral Lymph Node Metastasis Prediction

Contact Info

Product

Resources

About