Radiomics-based machine learning analysis and characterization of breast lesions with multiparametric diffusion-weighted MR

Enhancing radiomics features via a large language model for classifying benign and malignant breast tumors in mammography

BACKGROUND AND OBJECTIVES

Radiomics is widely used to assist in clinical decision-making, disease diagnosis, and treatment planning for various target organs, including the breast. Recent advances in large language models (LLMs) have helped enhance radiomics analysis.

MATERIALS AND METHODS

Herein, we sought to improve radiomics analysis by incorporating LLM-learned clinical knowledge, to classify benign and malignant tumors in breast mammography. We extracted radiomics features from the mammograms based on the region of interest and retained the features related to the target task. Using prompt engineering, we devised an input sequence that reflected the selected features and the target task. The input sequence was fed to the chosen LLM (LLaMA variant), which was fine-tuned using low-rank adaptation to enhance radiomics features. This was then evaluated on two mammogram datasets (VinDr-Mammo and INbreast) against conventional baselines.

RESULTS

The enhanced radiomics-based method performed better than baselines using conventional radiomics features tested on two mammogram datasets, achieving accuracies of 0.671 for the VinDr-Mammo dataset and 0.839 for the INbreast dataset. Conventional radiomics models require retraining from scratch for an unseen dataset using a new set of features. In contrast, the model developed in this study effectively reused the common features between the training and unseen datasets by explicitly linking feature names with feature values, leading to extensible learning across datasets. Our method performed better than the baseline method in this retraining setting using an unseen dataset.

CONCLUSIONS

Our method, one of the first to incorporate LLM into radiomics, has the potential to improve radiomics analysis.

Multi-Centered Pre-Treatment CT-Based Radiomics Features to Predict Locoregional Recurrence of Locally Advanced Esophageal Cancer After Definitive Chemoradiotherapy

Purpose: We constructed a prediction model to predict a 2-year locoregional recurrence based on the clinical features and radiomic features extracted from the machine learning method using computed tomography (CT) before definite chemoradiotherapy (dCRT) in locally advanced esophageal cancer. Patients and methods: A total of 264 patients (156 in Beijing, 87 in Tianjin, and 21 in Jiangsu) were included in this study. All those locally advanced esophageal cancer patients received definite radiotherapy and were randomly divided into five subgroups with a similar number and divided into training groups and validation groups by five cross-validations. The esophageal tumor and extratumoral esophagus were segmented to extract radiomic features from the gross tumor volume (GTV) drawn by radiation therapists before radiotherapy, and six clinical features associated with prognosis were added. T stage, N stage, M stage, total TNM stage, GTV, and GTVnd volume were included to construct a prediction model to predict the 2-year locoregional recurrence of patients after definitive radiotherapy. Results: A total of 264 patients were enrolled from August 2012 to April 2018, with a median age of 62 years and 81% were males. The 2-year locoregional recurrence rate was 52.6%, and the 2-year overall survival rate was 45.6%. About 66% of patients received concurrent chemotherapy. In total, we extracted 786 radiomic features from CT images and the Principal Component Analysis (PCA) method was used to screen out the maximum 30 features. Finally, the Support Vector Machine (SVM) method was used to construct the integrated prediction model combining radiomics and clinical features. In the five training groups for predicting locoregional recurrence, the mean value of C-index was 0.9841 (95%CI, 0.9809-0.9873), and in the five validation groups, the mean value was 0.744 (95%CI, 0.7437-0.7443). Conclusions: The integrated radiomics model could predict the 2-year locoregional recurrence after dCRT. The model showed promising results and could help guide treatment decisions by identifying high-risk patients and enabling strategies to prevent early recurrence.

MRI-based vector radiomics for predicting breast cancer HER2 status and its changes after neoadjuvant therapy

PURPOSE

To develop a novel MRI-based vector radiomic approach to predict breast cancer (BC) human epidermal growth factor receptor 2 (HER2) status (zero, low, and positive; task 1) and its changes after neoadjuvant therapy (NAT) (positive-to-positive, positive-to-negative, and positive-to-pathologic complete response; task 2).

MATERIALS AND METHODS

Both dynamic contrast-enhanced (DCE) MRI data and multi-b-value (MBV) diffusion-weighted imaging (DWI) data were acquired in BC patients at two centers. Vector-radiomic and conventional-radiomic features were extracted from both DCE-MRI and MBV-DWI. After feature selection, the following models were built using the retained features and logistic regression: vector model, conventional model, and combined model that integrates the vector-radiomic and conventional-radiomic features. The models' performances were quantified by the area under the receiver-operating characteristic curve (AUC).

RESULTS

The training/external test set (center 1/2) included 483/361 women. For task 1, the vector model (AUCs=0.73∼0.86) was superior to (p<.05) the conventional model (AUCs=0.68∼0.81), and the addition of vector-radiomic features to conventional-radiomic features yielded an incremental predictive value (AUCs=0.80∼0.90, p<.05). For task 2, the combined MBV-DWI model (AUCs=0.85∼0.89) performed better than (p<.05) the conventional MBV-DWI model (AUCs=0.73∼0.82). In addition, for the combined DCE-MRI model and the combined MBV-DWI model, the former (AUCs=0.85∼0.90) outperformed (p<.05) the latter (AUCs=0.80∼0.85) in task 1, whereas the latter (AUCs=0.85∼0.89) outperformed (p<.05) the former (AUCs=0.76∼0.81) in task 2. The above results are true for the training and external test sets.

CONCLUSIONS

MRI-based vector radiomics may predict BC HER2 status and its changes after NAT and provide significant incremental prediction over and above conventional radiomics.

Radiomics in Diagnosis, Grading, and Treatment Response Assessment of Soft Tissue Sarcomas: A Systematic Review and Meta-analysis

RATIONALE AND OBJECTIVES

To evaluate radiomics in soft tissue sarcomas (STSs) for diagnostic accuracy, grading, and treatment response assessment, with a focus on clinical relevance.

METHODS

In this diagnostic accuracy study, radiomics was applied using multiple MRI sequences and AI classifiers, with histopathological diagnosis as the reference standard. Statistical analysis involved meta-analysis, random-effects model, and Deeks' funnel plot asymmetry test.

RESULTS

Among 579 unique titles and abstracts, 24 articles were included in the systematic review, with 21 used for meta-analysis. Radiomics demonstrated a pooled sensitivity of 84% (95% CI: 80-87) and specificity of 63% (95% CI: 56-70), AUC of 0.93 for diagnosis, sensitivity of 84% (95% CI: 82-87) and specificity of 73% (95% CI: 68-77), AUC of 0.91 for grading, and sensitivity of 83% (95% CI: 67-94) and specificity of 67% (95% CI: 59-74), AUC of 0.87 for treatment response assessment.

CONCLUSION

Radiomics exhibits potential for accurate diagnosis, grading, and treatment response assessment in STSs, emphasizing the need for standardization and prospective trials.

CLINICAL RELEVANCE STATEMENT

Radiomics offers precise tools for STS diagnosis, grading, and treatment response assessment, with implications for optimizing patient care and treatment strategies in this complex malignancy.

Machine Learning Radiomics-Based Prediction of Non-sentinel Lymph Node Metastasis in Chinese Breast Cancer Patients with 1-2 Positive Sentinel Lymph Nodes: A Multicenter Study

RATIONALE AND OBJECTIVES

This study aimed to construct a machine learning radiomics-based model using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) images to evaluate non-sentinel lymph node (NSLN) metastasis in Chinese breast cancer (BC) patients who underwent total mastectomy (TM) and had 1-2 positive sentinel lymph nodes (SLNs).

MATERIALS AND METHODS

In total, 494 patients were retrospectively enrolled from two hospitals, and were divided into the training (n = 286), internal validation (n = 122), and external validation (n = 86) cohorts. Features were extracted from DCE-MRI images for each patient and screened. Six ML classifies were trained and the best classifier was evaluated to calculate radiomics (Rad)-scores. A combined model was developed based on Rad-scores and clinical risk factors, then the calibration, discrimination, reclassification, and clinical usefulness were evaluated.

RESULTS

14 radiomics features were ultimately selected. The random forest (RF) classifier showed the best performance, with the highest average area under the curve (AUC) of 0.833 in the validation cohorts. The combined model incorporating RF-based Rad-scores, tumor size, lymphovascular invasion, and proportion of positive SLNs resulted in the best discrimination ability, with AUCs of 0.903, 0.890, and 0.836 in the training, internal validation, and external validation cohorts, respectively. Furthermore, the combined model significantly improved the classification accuracy and clinical benefit for NSLN metastasis prediction.

CONCLUSION

A RF-based combined model using DCE-MRI images exhibited a promising performance for predicting NSLN metastasis in Chinese BC patients who underwent TM and had 1-2 positive SLNs, thereby aiding in individualized clinical treatment decisions.

A novel approach correlating pathologic complete response with digital pathology and radiomics in triple-negative breast cancer

This rapid communication highlights the correlations between digital pathology-whole slide imaging (WSI) and radiomics-magnetic resonance imaging (MRI) features in triple-negative breast cancer (TNBC) patients. The research collected 12 patients who had both core needle biopsy and MRI performed to evaluate pathologic complete response (pCR). The results showed that higher collagenous values in pathology data were correlated with more homogeneity, whereas higher tumor expression values in pathology data correlated with less homogeneity in the appearance of tumors on MRI by size zone non-uniformity normalized (SZNN). Higher myxoid values in pathology data are correlated with less similarity of gray-level non-uniformity (GLN) in tumor regions on MRIs, while higher immune values in WSIs correlated with the more joint distribution of smaller-size zones by small area low gray-level emphasis (SALGE) in the tumor regions on MRIs. Pathologic complete response (pCR) was associated with collagen, tumor, and myxoid expression in WSI and GLN and SZNN in radiomic features. The correlations of WSI and radiomic features may further our understanding of the TNBC tumoral microenvironment (TME) and could be used in the future to better tailor the use of neoadjuvant chemotherapy (NAC). This communication will focus on the post-NAC MRI features correlated with pCR and their association with WSI features from core needle biopsies.

MRI-based radiomics model to preoperatively predict mesenchymal transition subtype in high-grade serous ovarian cancer

AIM

To develop a magnetic resonance imaging (MRI)-based radiomics model for the preoperative identification of mesenchymal transition (MT) subtype in high-grade serous ovarian cancer (HGSOC).

MATERIALS AND METHODS

One hundred and eighty-nine patients with histopathologically confirmed HGSOC were enrolled retrospectively. Among the included patients, 55 patients were determined as the MT subtype and the remaining 134 were non-MT subtype. After extracting a total of 204 features from T2-weighted imaging (T2WI) and contrast-enhanced (CE)-T1WI images, the Mann-Whitney U-test, Spearman correlation test, and Boruta algorithm were adopted to select the optimal feature set. Three classifiers, including logistic regression (LR), support vector machine (SVM), and random forest (RF), were trained to develop radiomics models. The performance of established models was evaluated from three aspects: discrimination, calibration, and clinical utility.

RESULTS

Seven radiomics features relevant to MT subtypes were selected to build the radiomics models. The model based on the RF algorithm showed the best performance in predicting MT subtype, with areas under the curves (AUCs) of 0.866 (95 % confidence interval [CI]: 0.797-0.936) and 0.852 (95 % CI: 0.736-0.967) in the training and testing cohorts, respectively. The calibration curves, supported with Brier scores, indicated very good consistency between observation and prediction. Decision curve analysis (DCA) showed that the RF-based model could provide more net benefit, which suggested favorable utility in clinical application.

CONCLUSION

The RF-based radiomics model provided accurate identification of MT from the non-MT subtype and may help facilitate personalised management of HGSOC.

Machine learning and new insights for breast cancer diagnosis

Breast cancer (BC) is the most prominent form of cancer among females all over the world. The current methods of BC detection include X-ray mammography, ultrasound, computed tomography, magnetic resonance imaging, positron emission tomography and breast thermographic techniques. More recently, machine learning (ML) tools have been increasingly employed in diagnostic medicine for its high efficiency in detection and intervention. The subsequent imaging features and mathematical analyses can then be used to generate ML models, which stratify, differentiate and detect benign and malignant breast lesions. Given its marked advantages, radiomics is a frequently used tool in recent research and clinics. Artificial neural networks and deep learning (DL) are novel forms of ML that evaluate data using computer simulation of the human brain. DL directly processes unstructured information, such as images, sounds and language, and performs precise clinical image stratification, medical record analyses and tumour diagnosis. Herein, this review thoroughly summarizes prior investigations on the application of medical images for the detection and intervention of BC using radiomics, namely DL and ML. The aim was to provide guidance to scientists regarding the use of artificial intelligence and ML in research and the clinic.

Artificial intelligence in breast imaging: potentials and challenges

Breast cancer, which is the most common type of malignant tumor among humans, is a leading cause of death in females. Standard treatment strategies, including neoadjuvant chemotherapy, surgery, postoperative chemotherapy, targeted therapy, endocrine therapy, and radiotherapy, are tailored for individual patients. Such personalized therapies have tremendously reduced the threat of breast cancer in females. Furthermore, early imaging screening plays an important role in reducing the treatment cycle and improving breast cancer prognosis. The recent innovative revolution in artificial intelligence (AI) has aided radiologists in the early and accurate diagnosis of breast cancer. In this review, we introduce the necessity of incorporating AI into breast imaging and the applications of AI in mammography, ultrasonography, magnetic resonance imaging, and positron emission tomography/computed tomography based on published articles since 1994. Moreover, the challenges of AI in breast imaging are discussed.

Multimodality MRI radiomics analysis of TP53 mutations in triple negative breast cancer

Objectives

To explore the value of T1-weighted imaging (T1WI), T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) radiomics features reflecting TP53 mutations in patients with triple negative breast cancer (TNBC).

Study design

This retrospective study enrolled 91 patients with TNBC with TP53 testing (64 patients in the training cohort and 27 patients in the validation cohort). A total of 2832 radiomics features were extracted from the first phase of dynamic contrast-enhanced T1WI, T2WI and ADC maps. Analysis of variance (ANOVA) and the Kruskal-Wallis-test were used for feature selection. Then, linear discriminant analysis (LDA), multilayer perceptron (MLP), logistic regression (LR), LR with LASSO, decision tree (DT), naïve Bayes (NB), random forest (RF), and support vector machine (SVM) models were used for classification.

Results

The validation AUCs of the eight classifiers ranged from 0.74 (NB) to 0.85 (SVM). SVM attained the highest AUC (0.85) and diagnostic accuracy (0.82) of all tested models. The top 3 ranking features in the SVM model were T1-square-first order-skewness (coefficient: 1.735), T2-wavelet-LHH-GLCM-joint energy, and T2-wavelet-LHH-GLCM-inverse difference moment (coefficient: -0.654, -0.634).

Conclusions

Radiomics-based analysis with the SVM model is recommended for the detection of TP53 mutations in TNBC. Furthermore, T1WI- and T2WI-related features could be used as noninvasive biomarkers for predicting TP53 mutations.

A review of quantitative diffusion-weighted MR imaging for breast cancer: Towards noninvasive biomarker

Quantitative diffusion-weighted imaging (DWI) is an important adjunct to conventional breast MRI and shows promise as a noninvasive biomarker of breast cancer in multiple clinical scenarios, from the discrimination of benign and malignant lesions, prediction, and evaluation of treatment response to a prognostic assessment of breast cancer. Various quantitative parameters are derived from different DWI models based on special prior knowledge and assumptions, have different meanings, and are easy to confuse. In this review, we describe the quantitative parameters derived from conventional and advanced DWI models commonly used in breast cancer and summarize the promising clinical applications of these quantitative parameters. Although promising, it is still challenging for these quantitative parameters to become clinically useful noninvasive biomarkers in breast cancer, as multiple factors may result in variations in quantitative parameter measurements. Finally, we briefly describe some considerations regarding the factors that cause variations.

Radiomics features for assessing tumor-infiltrating lymphocytes correlate with molecular traits of triple-negative breast cancer

BACKGROUND

Tumor-infiltrating lymphocytes (TILs) have become a promising biomarker for assessing tumor immune microenvironment and predicting immunotherapy response. However, the assessment of TILs relies on invasive pathological slides.

METHODS

We retrospectively extracted radiomics features from magnetic resonance imaging (MRI) to develop a radiomic cohort of triple-negative breast cancer (TNBC) (n = 139), among which 116 patients underwent transcriptomic sequencing. This radiomic cohort was randomly divided into the training cohort (n = 98) and validation cohort (n = 41) to develop radiomic signatures to predict the level of TILs through a non-invasive method. Pathologically evaluated TILs in the H&E sections were set as the gold standard. Elastic net and logistic regression were utilized to perform radiomics feature selection and model training, respectively. Transcriptomics was utilized to infer the detailed composition of the tumor microenvironment and to validate the radiomic signatures.

RESULTS

We selected three radiomics features to develop a TILs-predicting radiomics model, which performed well in the validation cohort (AUC 0.790, 95% confidence interval (CI) 0.638-0.943). Further investigation with transcriptomics verified that tumors with high TILs predicted by radiomics (Rad-TILs) presented activated immune-related pathways, such as antigen processing and presentation, and immune checkpoints pathways. In addition, a hot immune microenvironment, including upregulated T cell infiltration gene signatures, cytokines, costimulators and major histocompatibility complexes (MHCs), as well as more CD8⁺ T cells, follicular helper T cells and memory B cells, was found in high Rad-TILs tumors.

CONCLUSIONS

Our study demonstrated the feasibility of radiomics model in predicting TILs status and provided a method to make the features interpretable, which will pave the way toward precision medicine for TNBC.

Machine learning for multi-parametric breast MRI: radiomics-based approaches for lesion classification

In the artificial intelligence era, machine learning (ML) techniques have gained more and more importance in the advanced analysis of medical images in several fields of modern medicine. Radiomics extracts a huge number of medical imaging features revealing key components of tumor phenotype that can be linked to genomic pathways. The multi-dimensional nature of radiomics requires highly accurate and reliable machine-learning methods to create predictive models for classification or therapy response assessment.

Multi-parametric breast magnetic resonance imaging (MRI) is routinely used for dense breast imaging as well for screening in high-risk patients and has shown its potential to improve clinical diagnosis of breast cancer. For this reason, the application of ML techniques to breast MRI, in particular to multi-parametric imaging, is rapidly expanding and enhancing both diagnostic and prognostic power. In this review we will focus on the recent literature related to the use of ML in multi-parametric breast MRI for tumor classification and differentiation of molecular subtypes. Indeed, at present, different models and approaches have been employed for this task, requiring a detailed description of the advantages and drawbacks of each technique and a general overview of their performances.

Prediction of the activity of Crohn's disease based on CT radiomics combined with machine learning models

PURPOSE

To investigate the value of a CT-based radiomics model in identification of Crohn's disease (CD) active phase and remission phase.

METHODS

CT images of 101 patients diagnosed with CD were retrospectively collected, which included 60 patients in active phase and 41 patients in remission phase. These patients were randomly divided into training group and test group at a ratio of 7 : 3. First, the lesion areas were manually delineated by the physician. Meanwhile, radiomics features were extracted from each lesion. Next, the features were selected by t-test and the least absolute shrinkage and selection operator regression algorithm. Then, several machine learning models including random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), logistic regression (LR) and K-nearest neighbor (KNN) algorithms were used to construct CD activity classification models respectively. Finally, the soft-voting mechanism was used to integrate algorithms with better effects to perform two classifications of data, and the receiver operating characteristic curves were applied to evaluate the diagnostic value of the models.

RESULTS

Both on the training set and the test set, AUC of the five machine learning classification models reached 0.85 or more. The ensemble soft-voting classifier obtained by using the combination of SVM, LR and KNN could better distinguish active CD from CD remission. For the test set, AUC was 0.938, and accuracy, sensitivity, and specificity were 0.903, 0.911, and 0.892, respectively.

CONCLUSION

This study demonstrated that the established radiomics model could objectively and effectively diagnose CD activity. The integrated approach has better diagnostic performance.