Multimodal Radiomics Research Articles

Automatic segmentation-based multi-modal radiomics analysis of US and MRI for predicting disease-free survival of breast cancer: a multicenter study

BackgroundSeveral studies have confirmed the potential value of applying radiomics to predict prognosis of breast cancer. However, the tumor segmentation in these studies depended on delineation or annotation of breast cancer by radiologist, which is often laborious, tedious, and vulnerable to inter- and intra-observer variability. Automatic segmentation is expected to overcome this difficulty. Herein, we aim to investigate the value of automatic segmentation-based multi-modal radiomics signature and magnetic resonance imaging (MRI) features in predicting disease-free survival (DFS) of patients diagnosed with invasive breast cancer.MethodsThis retrospective multicenter study included a total of 643 female patients with invasive breast cancer who underwent preoperative ultrasound (US) and MRI for prognostic analysis. Data (n = 480) from center 1 were divided into training and internal testing sets, while data (n = 163) from centers 2 and 3 were analyzed as the external testing set. We developed automatic segmentation frameworks for tumor segmentation by deep learning. Then, Least absolute shrinkage and selection operator Cox regression was used to select features to construct radiomics signature, and corresponding radiomics score (Rad-score) was calculated. Finally, six models for predicting DFS were constructed by using Cox regression and assessed in terms of discrimination, calibration, and clinical usefulness.ResultsThe multi-modal radiomics signature combining intra- and peri-tumoral radiomics signatures of US and MRI achieved a higher C-index in the internal (0.734) and external (0.708) testing sets than most other radiomics signatures in predicting DFS, and successfully stratified patients into low- and high-risk groups. The multi-modal clinical imaging model combining the multi-modal Rad-score and clinical traditional MRI model-score resulted in a higher C-index (0.795) than other models in the external testing set, and it had a better calibration and higher clinical benefit.ConclusionsThis study demonstrates that the multi-modal radiomics signature derived from automatic segmentations of US and MRI is a promising risk stratification biomarker for breast cancer, and highlights that the appropriate combination of multi-modal radiomics signature, clinical characteristics, and MRI feature can improve performance of individualized DFS prediction, which might assist in guiding decision-making related to breast cancer.

Multimodal radiomics and deep learning models for predicting early femoral head deformity in LCPD

PurposeTo develop a predictive model combining clinical, radiomic, and deep learning features based on X-ray and MRI to identify risk factors for early femoral head deformity in Legg-Calvé-Perthes disease (LCPD). MethodsThis study involved 152 patients diagnosed with early unilateral LCPD across two centers between January 2013 and December 2023, and included an independent external validation set to assess generalizability. Four machine learning methods, including logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost), were employed to develop radiomics deep learning signatures. The clinical-radiomics model (Clinic + Rad), clinical-deep learning model (Clinic + DL), and clinical-radiomics-deep learning model (Clinic + Rad + DL) were developed by integrating radiomics deep learning signatures with clinical variables. The best model, integrated into a nomogram for clinical application, was evaluated using the area under the receiver operating characteristic curve (AUC). ResultsAmong the four machine learning methods, XGBoost demonstrated superior performance in our patient dataset: radiomic (Rad) model (AUC, 0.786) and deep learning (DL) model (AUC, 0.803). Clinical variables such as age at onset and JIC classification were associated with early femoral head deformity (p < 0.05). The combined model incorporating clinical, radiomic, and deep learning signatures demonstrated better predictive ability (AUC, 0.853). The nomogram can assist clinicians in effectively assessing the risk of early femoral head deformity. ConclusionThe Clinic + Rad + DL integrated model may be beneficial for prognostic assessment of early LCPD femoral head deformity, which is crucial for tailoring personalized treatment strategies for individual patients.

Enhancing Preoperative Diagnosis of Subscapular Muscle Injuries with Shoulder MRI-based Multimodal Radiomics

PurposeRotator cuff injury is a common ailment in the musculoskeletal system, with the subscapularis muscle being the largest and most robust muscle of the rotator cuff. The occurrence of subscapularis muscle tears is more frequent than previously reported. The main objective of this research is to harness the power of artificial intelligence to enhance the precision in diagnosing subscapularis muscle injuries via magnetic resonance imaging of the shoulder joint, prior to surgical intervention. This study seeks to integrate advanced artificial intelligence algorithms to analyze magnetic resonance imaging data, aiming to provide more accurate preoperative assessments, which can potentially lead to better surgical outcomes and patient care and promote technological progress in the field of medical imaging analysis. MethodThis is a multicenter study that involves 324 patients from a major medical center serving as both the training and testing groups, with an additional 60 patients from two other medical centers comprising the verifying group. The imaging protocol for all these subjects included a series of shoulder magnetic resonance imaging scans: T1-weighted coronal sequences, T2-weighted coronal, axial, and sagittal images. These comprehensive imaging modalities were utilized to thoroughly examine the shoulder joint's anatomical details and to detect any signs of subscapularis muscle damage. To enhance the diagnostic accuracy before surgical procedures, radiomic analysis was employed. This technique involves the extraction of a multitude of quantitative features from the magnetic resonance imaging, which can provide a more nuanced and data-driven approach to identifying subscapularis muscle injuries. The integration of radiomics in this study aims to offer a more precise preoperative assessment, potentially leading to improved surgical planning and patient outcomes. ResultIn the course of this study, a comprehensive extraction of 1197 radiomic features was performed for each imaging modality of every patient. The coronal T1-weighted modality, when assessed within the internal verifying cohort, delivered a diagnostic accuracy of 0.766, coupled with an AUC of 0.803. In the case of the T2-weighted modality, the coronal planes exhibited a diagnostic accuracy of 0.781 and an AUC of 0.844. The axial T2-weighted images recorded an accuracy of 0.719 and an AUC of 0.761, while the sagittal T2-weighted images scored an accuracy of 0.766 and an AUC of 0.821. The amalgamation of these imaging techniques through a multimodal strategy markedly enhanced the accuracy to 0.828, with an AUC of 0.916 for the internal verifying group. The diagnostic performance of the coronal T1-weighted modality in the external verifying cohort yielded an accuracy of 0.833, with an area under the curve (AUC) of 0.819. For the T2-weighted modality, the coronal imaging demonstrated an accuracy of 0.767 and an AUC of 0.794. The axial T2-weighted images had an accuracy of 0.783 and an AUC of 0.797, while the sagittal T2-weighted images achieved an accuracy of 0.833 and an AUC of 0.800. When combining the modalities, the multimodal approach significantly improved the accuracy to 0.867, with an AUC of 0.803 in the external verifying group, indicating a robust diagnostic capability. ConclusionOur study demonstrates that the application of multimodal radiomic techniques to shoulder magnetic resonance imaging significantly enhances the precision of preoperative diagnosis for subscapularis muscle injuries. This approach leverages the comprehensive data provided by various magnetic resonance imaging modalities to offer a more detailed and accurate assessment, which is crucial for surgical planning and patient care.

Utilizing multimodal radiomics technology from cervical MRI for diagnosis of cervical spinal cord injury and spinal cord concussion

The primary aim of this study is to assess the viability of employing multimodal radiomics techniques for distinguishing between cervical spinal cord injury and spinal cord concussion in cervical magnetic resonance imaging. This is a multicenter study involving 288 patients from a major medical center as the training group, and 75 patients from two other medical centers as the testing group. Data regarding the presence of spinal cord injury symptoms and their recovery status within 72 h were documented. These patients underwent sagittal T1-weighted and T2-weighted imaging using cervical magnetic resonance imaging. Radiomics techniques are used to help diagnose whether these patients have cervical spinal cord injury or spinal cord concussion. 1197 radiomics features were extracted for each modality of each patient. The accuracy of T1 modal in testing group is 0.773, AUC is 0.799. The accuracy of T2 modal in testing group is 0.707, AUC is 0.813. The accuracy of T1 + T2 modal in testing group is 0.800, AUC is 0.840. Our research indicates that multimodal radiomics techniques utilizing cervical magnetic resonance imaging can effectively diagnose the presence of cervical spinal cord injury or spinal cord concussion.

134 Multimodal radiomics predicts pathological response of esophageal cancer after neoadjuvant therapy

134 Multimodal radiomics predicts pathological response of esophageal cancer after neoadjuvant therapy

Radiomics in oncology: A mini-review of principles, applications and challenges

Radiomics, a quantitative approach to medical imaging, employs computational methods to extract features from the images, revealing hidden characteristics of specific regions. This emerging field leverages advanced techniques to analyze a spectrum of features from modalities, including computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) scans, aiming to decode tissue characteristics, disease progression, and treatment responses. The radiomics workflow integrates image acquisition, segmentation, feature selection, and data integration, utilizing advanced techniques such as deep learning, machine learning, and data mining. Radiomics demonstrates considerable potential in cancer detection and management, exhibiting high sensitivity and specificity in distinguishing between benign and malignant tumors and predicting outcomes. However, challenges such as imaging protocol variability, overfitting, and standardization requirements impede its broad clinical adoption. Innovations in multi-modal radiomics, deep learning, and genomics integration strive to mitigate these constraints. This review elucidates radiomics’ capabilities, current applications, benefits, challenges, and future directions in oncology.

Predicting Outcome of Patients With Cerebral Hemorrhage Using a Computed Tomography-Based Interpretable Radiomics Model: A Multicenter Study.

The aim of this study was to develop and validate an interpretable and highly generalizable multimodal radiomics model for predicting the prognosis of patients with cerebral hemorrhage. This retrospective study involved 237 patients with cerebral hemorrhage from 3 medical centers, of which a training cohort of 186 patients (medical center 1) was selected and 51 patients from medical center 2 and medical center 3 were used as an external testing cohort. A total of 1762 radiomics features were extracted from nonenhanced computed tomography using Pyradiomics, and the relevant macroscopic imaging features and clinical factors were evaluated by 2 experienced radiologists. A radiomics model was established based on radiomics features using the random forest algorithm, and a radiomics-clinical model was further trained by combining radiomics features, clinical factors, and macroscopic imaging features. The performance of the models was evaluated using area under the curve (AUC), sensitivity, specificity, and calibration curves. Additionally, a novel SHAP (SHAPley Additive exPlanations) method was used to provide quantitative interpretability analysis for the optimal model. The radiomics-clinical model demonstrated superior predictive performance overall, with an AUC of 0.88 (95% confidence interval, 0.76-0.95; P < 0.01). Compared with the radiomics model (AUC, 0.85; 95% confidence interval, 0.72-0.94; P < 0.01), there was a 0.03 improvement in AUC. Furthermore, SHAP analysis revealed that the fusion features, rad score and clinical rad score, made significant contributions to the model's decision-making process. Both proposed prognostic models for cerebral hemorrhage demonstrated high predictive levels, and the addition of macroscopic imaging features effectively improved the prognostic ability of the radiomics-clinical model. The radiomics-clinical model provides a higher level of predictive performance and model decision-making basis for the risk prognosis of cerebral hemorrhage.

Preoperative prediction of HER-2 expression status in breast cancer based on MRI radiomics model

Objective: This study aims to explore the predictive value of T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC), and early-delayed phases enhanced magnetic resonance imaging (DCE-MRI) radiomics prediction model in determining human epidermal growth factor receptor 2 status in breast cancer. Methods: A retrospective study was conducted, involving 187 patients with confirmed breast cancer by postsurgical pathology at Zhenjiang First People's Hospital during January 2021 and May 2023. Immunohistochemistry or fluorescence in situ hybridization was used to determine the HER-2 status of these patients, with 48 cases classified as HER-2 positive and 139 cases as HER-2 negative. The training set was used to construct the prediction models and the validation set was used to verify the prediction models. Layers of T2WI, ADC, and early-delayed phase DCE-MRI images were used to delineate the volumeof interest and 960 radiomic features were extracted from each case using Pyradiomic. After screening and dimensionality reduction by intraclass correlation coefficient, Pearson correlation analysis, least absolute shrinkage, and selection operator, the radiomics labels were established. Logistic regression analysis was used to construct the T2WI radiomics model, ADC radiomics model, DCE-2 radiomics model, DCE-6 radiomics model, and the joint sequence radiomics model to predict the HER-2 expression status of breast cancer, respectively. Based on the clinical, pathological, and MRI image characteristics of patients, univariate and multivariate logistic regression analysis wasused to construct a clinicopathological MRI feature model. The radscore of every patient and the clinicopathological MRI features which were statistically significant after screening were used to construct a nomogram model. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of each model and the decision curve analysis wasused to evaluate the clinical usefulness. Results: The T2WI, ADC, DCE-2, DCE-6, and joint sequence radiomics models, the clinicopathological MRI feature model, and the nomogram model were successfully constructed to predict the expression status of HER-2 in breast cancer. ROC analysis showed that in the training set and validation set, the areas under the curve (AUC) of the T2WI radiomics model were 0.797 and 0.760, of the ADC radiomics model were 0.776 and 0.634, of the DCE-2 radiomics model were 0.804 and 0.759, of the DCE-6 radiomics model were 0.869 and 0.798, of the combined sequence radiomics model were 0.908 and 0.847, of the clinicopathological MRI feature model were 0.703 and 0.693, and of the nomogram model were 0.938 and 0.859, respectively. In the training set, the combined sequence radiomics model outperformed the clinicopathological features model (P＜0.001). In the training and validation sets, the nomogram outperformed the clinicopathological features model (P＜0.05). In addition, the diagnostic performance of the nomogram was better than that of the four single-modality radiomics models in the training cohort (P＜0.05) and was better than that of DCE-2 and ADC models in the validation cohort (P＜0.05). Decision curve analysis indicated that the value of individualized prediction models was higher than clinical and pathological prediction models in clinical practice. The calibration curve showed that the multimodal radiomics model had a high consistency with the actual results in predicting HER-2 expression. Conclusions: T2WI, ADC and early-delayed phase DCE-MRI imaging histology models for HER-2 expression status in breast cancer are expected to provide a non-invasive virtual pathological basis for decision-making on preoperative neoadjuvant regimens in breast cancer.

Predicting long-term outcomes for acute ischemic stroke using multi-model MRI radiomics and clinical variables.

The objective of this study was to create and validate a novel prediction model that incorporated both multi-modal radiomics features and multi-clinical features, with the aim of accurately identifying acute ischemic stroke (AIS) patients who faced a higher risk of poor outcomes. A cohort of 461 patients diagnosed with AIS from four centers was divided into a training cohort and a validation cohort. Radiomics features were extracted and selected from diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) images to create a radiomic signature. Prediction models were developed using multi-clinical and selected radiomics features from DWI and ADC. A total of 49 radiomics features were selected from DWI and ADC images by the least absolute shrinkage and selection operator (LASSO). Additionally, 20 variables were collected as multi-clinical features. In terms of predicting poor outcomes in validation set, the area under the curve (AUC) was 0.727 for the DWI radiomics model, 0.821 for the ADC radiomics model, 0.825 for the DWI + ADC radiomics model, and 0.808 for the multi-clinical model. Furthermore, a prediction model was built using all selected features, the AUC for predicting poor outcomes increased to 0.86. Radiomics features extracted from DWI and ADC images can serve as valuable biomarkers for predicting poor clinical outcomes in patients with AIS. Furthermore, when these radiomics features were combined with multi-clinical features, the predictive performance was enhanced. The prediction model has the potential to provide guidance for tailoring rehabilitation therapies based on individual patient risks for poor outcomes.

A machine learning approach using 18F-FDG PET and enhanced CT scan-based radiomics combined with clinical model to predict pathological complete response in ESCC patients after neoadjuvant chemoradiotherapy and anti-PD-1 inhibitors.

We aim to evaluate the value of an integrated multimodal radiomics with machine learning model to predict the pathological complete response (pCR) of primary tumor in a prospective cohort of esophageal squamous cell carcinoma (ESCC) treated with neoadjuvant chemoradiotherapy (nCRT) and anti-PD-1 inhibitors. Clinical information of 126 ESCC patients were included for analysis. Radiomics features were extracted from 18F-FDG PET and enhanced plan CT images. Four machine learning algorithms, including SVM (Support Vector Machine), Random Forest (RF), and eXtreme Gradient Boosting (XGB) and logistic regression (LR), were applied using k-fold cross-validation to predict pCR after nCRT. The predictive ability of the models was assessed using receiver operating characteristics (ROC) curve analysis. A total of 842 features were extracted. Among the four machine learning algorithms, SVM achieved the most promising performance on the test set for PET(AUC:0.775), CT (AUC:0.710) and clinical model (AUC:0.722). For all combinations of various modalities-based models, the combination model of 18 F-FDG PET, CT and clinical features with SVM machine learning had the highest AUC of 0.852 in the test set when compared to single-modality models in various algorithms. The other combined models had AUC ranged 0.716 to 0.775. Machine learning models utilizing radiomics features from 18F-FDG PET and enhanced plan CT exhibit promising performance in predicting pCR in ESCC after nCRT and anti-PD-1 inhibitors. The fusion of features from multiple modalities radiomics and clinical features enhances the better predictive performance compared to using a single modality alone.

CNN-based multi-modal radiomics analysis of pseudo-CT utilization in MRI-only brain stereotactic radiotherapy: a feasibility study

BackgroundPseudo-computed tomography (pCT) quality is a crucial issue in magnetic resonance image (MRI)-only brain stereotactic radiotherapy (SRT), so this study systematically evaluated it from the multi-modal radiomics perspective.Methods34 cases (< 30 cm³) were retrospectively included (2021.9-2022.10). For each case, both CT and MRI scans were performed at simulation, and pCT was generated by a convolutional neural network (CNN) from planning MRI. Conformal arc or volumetric modulated arc technique was used to optimize the dose distribution. The SRT dose was compared between pCT and planning CT with dose volume histogram (DVH) metrics and gamma index. Wilcoxon test and Spearman analysis were used to identify key factors associated with dose deviations. Additionally, original image features were extracted for radiomic analysis. Tumor control probability (TCP) and normal tissue complication probability (NTCP) were employed for efficacy evaluation.ResultsThere was no significant difference between pCT and planning CT except for radiomics. The mean value of Hounsfield unit of the planning CT was slightly higher than that of pCT. The Gadolinium-based agents in planning MRI could increase DVH metrics deviation slightly. The median local gamma passing rates (1%/1 mm) between planning CTs and pCTs (non-contrast) was 92.6% (range 63.5–99.6%). Also, differences were observed in more than 85% of original radiomic features. The mean absolute deviation in TCP was 0.03%, and the NTCP difference was below 0.02%, except for the normal brain, which had a 0.16% difference. In addition, the number of SRT fractions and lesions, and lesion morphology could influence dose deviation.ConclusionsThis is the first multi-modal radiomics analysis of CNN-based pCT from planning MRI for SRT of small brain lesions, covering dosiomics and radiomics. The findings suggest the potential of pCT in SRT plan design and efficacy prediction, but caution needs to be taken for radiomic analysis.

Diagnosing postoperative lymph node metastasis in thyroid cancer with multimodal radiomics and clinical features.

This study aims to evaluate the diagnostic value of texture analysis for lymph node metastasis after thyroid cancer surgery. We retrospectively analyzed patients who underwent positron emission tomography/computed tomography (PET/CT) examination before 131I treatment at Shanghai Tenth People's Hospital between 2017 and 2020. Clinical follow-up results were used as the criterion for determining the presence of lymph node metastasis. The study included 119 patients, who were then randomly divided into training and test groups in a 7:3 ratio. Regions of interest were identified, and radiomics features were extracted using LIFEx 7.3.0. Mann-Whitney U test and LASSO regression were employed to screen radiomics parameters for modeling. Subsequently, a nomogram model was built by combining radscore and clinical features. SPSS 26.0 software was utilized for statistical analysis, and p < 0.05 was considered statistically significant. Follow-up confirmed 54 patients with thyroid cancer lymph node metastasis and 65 patients in the non-metastasis group. A total of 119 lymph nodes were delineated. For each lesion, 164 CT texture features and 164 PET texture features were extracted, and 107 significant parameters were identified, including 16 CT texture parameters and 91 PET texture parameters. After screening, 3 CT parameters, 4 PET parameters and 12 PET/CT parameters were selected to establish three radiomic models. The AUC values were as follows: AUC (CT) = 0.730, AUC (PET) = 0.759 and AUC (PET/CT) = 0.864. We then combined clinical features and radscore to construct a nomogram, resulting in a C-index of 0.915 in the training group. In the test group, the C-index was confirmed to be 0.868. Radiomics may enhance the diagnostic efficiency of lymph node metastases after thyroid cancer surgery and could potentially assist clinicians in future diagnoses. The developed nomogram, which combines radiomics and clinical features, offers relatively high accuracy in helping clinicians assess the risk of metastasis in thyroid patients after surgery.

A Novel Machine Learning Approach for Predicting Neoadjuvant Chemotherapy Response in Breast Cancer: Integration of Multimodal Radiomics With Clinical and Molecular Subtype Markers

A Novel Machine Learning Approach for Predicting Neoadjuvant Chemotherapy Response in Breast Cancer: Integration of Multimodal Radiomics With Clinical and Molecular Subtype Markers

Multimodal radiomics based on 18F-Prostate-specific membrane antigen-1007 PET/CT and multiparametric MRI for prostate cancer extracapsular extension prediction.

To compare the performance of the multiparametric magnetic resonance imaging (mpMRI) radiomics and 18F-Prostate-specific membrane antigen (PSMA)-1007 PET/CT radiomics model in diagnosing extracapsular extension (EPE) in prostate cancer (PCa), and to evaluate the performance of a multimodal radiomics model combining mpMRI and PET/CT in predicting EPE. We included 197 patients with PCa who underwent preoperative mpMRI and PET/CT before surgery. mpMRI and PET/CT images were segmented to delineate the regions of interest and extract radiomics features. PET/CT, mpMRI, and multimodal radiomics models were constructed based on maximum correlation, minimum redundancy, and logistic regression analyses. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and indices derived from the confusion matrix. AUC values for the mpMRI, PET/CT, and multimodal radiomics models were 0.85 (95% CI, 0.78-0.90), 0.73 (0.64-0.80), and 0.83 (0.75-0.89), respectively, in the training cohort and 0.74 (0.61-0.85), 0.62 (0.48-0.74), and 0.77 (0.64-0.87), respectively, in the testing cohort. The net reclassification improvement demonstrated that the mpMRI radiomics model outperformed the PET/CT one in predicting EPE, with better clinical benefits. The multimodal radiomics model performed better than the single PET/CT radiomics model (P < .05). The mpMRI and 18F-PSMA-PET/CT combination enhanced the predictive power of EPE in patients with PCa. The multimodal radiomics model will become a reliable and robust tool to assist urologists and radiologists in making preoperative decisions. This study presents the first application of multimodal radiomics based on PET/CT and MRI for predicting EPE.

CT-based nomogram for early identification of T790M resistance in metastatic non-small cell lung cancer before first-line epidermal growth factor receptor-tyrosine kinase inhibitors therapy

BackgroundTo evaluate the value of computed tomography (CT) radiomics in predicting the risk of developing epidermal growth factor receptor (EGFR) T790M resistance mutation for metastatic non-small lung cancer (NSCLC) patients before first-line EGFR-tyrosine kinase inhibitors (EGFR-TKIs) therapy.MethodsA total of 162 metastatic NSCLC patients were recruited and split into training and testing cohort. Radiomics features were extracted from tumor lesions on nonenhanced CT (NECT) and contrast-enhanced CT (CECT). Radiomics score (rad-score) of two CT scans was calculated respectively. A nomogram combining two CT scans was developed to evaluate T790M resistance within up to 14 months. Patients were followed up to calculate the time of T790M occurrence. Models were evaluated by area under the curve at receiver operating characteristic analysis (ROC-AUC), calibration curve, and decision curve analysis (DCA). The association of the nomogram with the time of T790M occurrence was evaluated by Kaplan–Meier survival analysis.ResultsThe nomogram constructed with the rad-score of NECT and CECT for predicting T790M resistance within 14 months achieved the highest ROC-AUCs of 0.828 and 0.853 in training and testing cohorts, respectively. The DCA showed that the nomogram was clinically useful. The Kaplan–Meier analysis showed that the occurrence time of T790M difference between the high- and low-risk groups distinguished by the rad-score was significant (p < 0.001).ConclusionsThe CT-based radiomics signature may provide prognostic information and improve pretreatment risk stratification in EGFR NSCLC patients before EGFR-TKIs therapy. The multimodal radiomics nomogram further improved the capability.Relevance statementRadiomics based on NECT and CECT images can effectively identify and stratify the risk of T790M resistance before the first-line TKIs treatment in metastatic non-small cell lung cancer patients.Key points• Early identification of the risk of T790M resistance before TKIs treatment is clinically relevant.• Multimodel radiomics nomogram holds potential to be a diagnostic tool.• It provided an imaging surrogate for identifying the pretreatment risk of T790M.Graphical

A diagnostic approach integrated multimodal radiomics with machine learning models based on lumbar spine CT and X-ray for osteoporosis.

The aim of this analysis is to construct a combined model that integrates radiomics, clinical risk factors, and machine learning algorithms to diagnose osteoporosis in patients and explore its potential in clinical applications. A retrospective analysis was conducted on 616 lumbar spine. Radiomics features were extracted from the computed tomography (CT) scans and anteroposterior and lateral X-ray images of the lumbar spine. Logistic regression (LR), support vector machine (SVM), and random forest (RF) algorithms were used to construct radiomics models. The receiver operating characteristic curve (ROC) was employed to select the best-performing model. Clinical risk factors were identified through univariate logistic regression analysis (ULRA) and multivariate logistic regression analysis (MLRA) and utilized to develop a clinical model. A combined model was then created by merging radiomics and clinical risk factors. The performance of the models was evaluated using ROC curve analysis, and the clinical value of the models was assessed using decision curve analysis (DCA). A total of 4858 radiomics features were extracted. Among the radiomics models, the SVM model demonstrated the optimal diagnostic capabilities and accuracy, with an area under the curve (AUC) of 0.958 (0.9405-0.9762) in the training cohort and 0.907 (0.8648-0.9492) in the test cohort. Furthermore, the combined model exhibited an AUC of 0.959 (0.9412-0.9763) in the training cohort and 0.910 (0.8690-0.9506) in the test cohort. The combined model displayed outstanding ability in diagnosing osteoporosis, providing a safe and efficient method for clinical decision-making.

Study of radiochemotherapy decision-making for young high-risk low-grade glioma patients using a macroscopic and microscopic combined radiomics model.

As a few types of glioma, young high-risk low-grade gliomas (HRLGGs) have higher requirements for postoperative quality of life. Although adjuvant chemotherapy with delayed radiotherapy is the first treatment strategy for HRLGGs, not all HRLGGs benefit from it. Accurate assessment of chemosensitivity in HRLGGs is vital for making treatment choices. This study developed a multimodal fusion radiomics (MFR) model to support radiochemotherapy decision-making for HRLGGs. A MFR model combining macroscopic MRI and microscopic pathological images was proposed. Multiscale features including macroscopic tumor structure and microscopic histological layer and nuclear information were grabbed by unique paradigm, respectively. Then, these features were adaptively incorporated into the MFR model through attention mechanism to predict the chemosensitivity of temozolomide (TMZ) by means of objective response rate and progression free survival (PFS). Macroscopic tumor texture complexity and microscopic nuclear size showed significant statistical differences (p < 0.001) between sensitivity and insensitivity groups. The MFR model achieved stable prediction results, with an area under the curve of 0.950 (95% CI: 0.942-0.958), sensitivity of 0.833 (95% CI: 0.780-0.848), specificity of 0.929 (95% CI: 0.914-0.936), positive predictive value of 0.833 (95% CI: 0.811-0.860), and negative predictive value of 0.929 (95% CI: 0.914-0.934). The predictive efficacy of MFR was significantly higher than that of the reported molecular markers (p < 0.001). MFR was also demonstrated to be a predictor of PFS. A MFR model including radiomics and pathological features predicts accurately the response postoperative TMZ treatment. Our MFR model could identify young high-risk low-grade glioma patients who can have the most benefit from postoperative upfront temozolomide (TMZ) treatment. • Multimodal radiomics is proposed to support the radiochemotherapy of glioma. • Some macro and micro image markers related to tumor chemotherapy sensitivity are revealed. • The proposed model surpasses reported molecular markers, with a promising area under the curve (AUC) of 0.95.

A clinical-radiomics nomogram based on multimodal ultrasound for predicting the malignancy risk in solid hypoechoic breast lesions.

In routine clinical examinations, solid hypoechoic breast lesions are frequently encountered, but accurately distinguishing them poses a challenge. This study proposed a clinical-radiomics nomogram based on multimodal ultrasound that enhances the diagnostic accuracy for solid hypoechoic breast lesions. This retrospective study analyzed ultrasound strain elastography (SE) and automated breast volume scanner images (ABVS) of 423 solid hypoechoic breast lesions from 423 female patients in our hospital between August 2019 and May 2022. They were assigned to the training (n=296) and validation (n=127) groups in a 7:3 ratio by generating random numbers. Radiomics features were extracted and screened from ABVS and SE images, followed by the calculation of the radiomics score (Radscore) based on these features. Subsequently, a nomogram was constructed through multivariate logistic regression to assess the malignancy risk in breast lesions by combining Radscore with Breast Imaging Reporting and Data System (BI-RADS) scores and clinical risk factors associated with breast malignant lesions. The diagnostic performance, calibration performance, and clinical usefulness of the nomogram were assessed by the area under the curve (AUC) of the receiver operating characteristic curve, the calibration curve, and the decision analysis curve, respectively. The diagnostic performance of the nomogram is significantly superior to that of both the clinical diagnostic model (BI-RADS model) and the multimodal radiomics model (SE+ABVS radiomics model) in training (AUC: 0.972 vs 0.930 vs 0.941) and validation group (AUC:0.964 vs 0.916 vs 0.933). In addition, the nomogram also exhibited a favorable goodness-of-fit and could lead to greater net benefits for patients. The nomogram enables a more effective assessment of the malignancy risk of solid hypoechoic breast lesions; therefore, it can serve as a new and efficient diagnostic tool for clinical diagnosis.

Can multi-modal radiomics using pretreatment ultrasound and tomosynthesis predict response to neoadjuvant systemic treatment in breast cancer?

ObjectivesResponse assessment to neoadjuvant systemic treatment (NAST) to guide individualized treatment in breast cancer is a clinical research priority. We aimed to develop an intelligent algorithm using multi-modal pretreatment ultrasound and tomosynthesis radiomics features in addition to clinical variables to predict pathologic complete response (pCR) prior to the initiation of therapy.MethodsWe used retrospective data on patients who underwent ultrasound and tomosynthesis before starting NAST. We developed a support vector machine algorithm using pretreatment ultrasound and tomosynthesis radiomics features in addition to patient and tumor variables to predict pCR status (ypT0 and ypN0). Findings were compared to the histopathologic evaluation of the surgical specimen. The main outcome measures were area under the curve (AUC) and false-negative rate (FNR).ResultsWe included 720 patients, 504 in the development set and 216 in the validation set. Median age was 51.6 years and 33.6% (242 of 720) achieved pCR. The addition of radiomics features significantly improved the performance of the algorithm (AUC 0.72 to 0.81; p = 0.007). The FNR of the multi-modal radiomics and clinical algorithm was 6.7% (10 of 150 with missed residual cancer). Surface/volume ratio at tomosynthesis and peritumoral entropy characteristics at ultrasound were the most relevant radiomics. Hormonal receptors and HER-2 status were the most important clinical predictors.ConclusionA multi-modal machine learning algorithm with pretreatment clinical, ultrasound, and tomosynthesis radiomics features may aid in predicting residual cancer after NAST. Pending prospective validation, this may facilitate individually tailored NAST regimens.Clinical relevance statementMulti-modal radiomics using pretreatment ultrasound and tomosynthesis showed significant improvement in assessing response to NAST compared to an algorithm using clinical variables only. Further prospective validation of our findings seems warranted to enable individualized predictions of NAST outcomes.Key Points• We proposed a multi-modal machine learning algorithm with pretreatment clinical, ultrasound, and tomosynthesis radiomics features to predict response to neoadjuvant breast cancer treatment.• Compared with the clinical algorithm, the AUC of this integrative algorithm is significantly higher.• Used prior to the initiative of therapy, our algorithm can identify patients who will experience pathologic complete response following neoadjuvant therapy with a high negative predictive value.

Automatic statistical diagnosis of COVID-19 based on multi-modal CT feature extraction

Background and purposeComputed tomography (CT) is highly sensitive to lung-related abnormalities as a non-invasive method and has become an essential tool for screening and diagnosing Coronavirus disease 2019 (COVID-19). To reduce the stress of work for physicians and speed up diagnosis, we propose a novel automatic diagnosis pipeline for COVID-19 based on high-dimensional radiomic features extracted from multimodal CT scans (multi-geometric and multiscale). Materials and methodsThere are 746 CT scans involved in this study, where 349 CT scans are COVID-19 positive and 397 CT scans are COVID-19 negative. All of them are from the public dataset. We first construct a transfer learning-based auto-segmentation model with a morphological post-processing block to improve the lung region segmentation. Then the radiomics feature extraction is guided by the proposed multi-modal CT scans strategy. In addition, our automatic diagnosis pipeline is driven by a well-designed loss function. We also explain the diagnosis capability from the related theory of linear subspace spanned by multi-modal radiomics features. ResultsUnder the 10-fold cross-validation strategy, our approach can achieve an improvement in diagnostic performance of 5. 77%, 7. 78%, 7. 74%, 7. 78%, 7. 45% compared to the radiomic features extracted from the original CT scans, and diagnosis performance is promoted to 91.53%, 86.46%, 86.47%, 86.46%, 86.95% in terms of AUC, Acc, F1, Recall and Precision in public datasets. ConclusionsWe demonstrate a statistically significant improvement of the proposed statistical learning method compared to the state-of-the-art machine learning-based diagnosis approaches. Thanks to theoretical support and excellent diagnostic performance, our method can be deployed in clinical auxiliary diagnosis, releasing the overstretched medical resources.