MRI Radiomics Research Articles

A novel radiomics based on multi-parametric magnetic resonance imaging for predicting Ki-67 expression in rectal cancer: a multicenter study

BackgroundTo explore the value of multiparametric MRI markers for preoperative prediction of Ki-67 expression among patients with rectal cancer.MethodsData from 259 patients with postoperative pathological confirmation of rectal adenocarcinoma who had received enhanced MRI and Ki-67 detection was divided into 4 cohorts: training (139 cases), internal validation (in-valid, 60 cases), and external validation (ex-valid, 60 cases) cohorts. The patients were divided into low and high Ki-67 expression groups. In the training cohort, DWI, T2WI, and contrast enhancement T1WI (CE-T1) sequence radiomics features were extracted from MRI images. Radiomics marker scores and regression coefficient were then calculated for data fitting to construct a radscore model. Subsequently, clinical features with statistical significance were selected to construct a combined model for preoperative individualized prediction of rectal cancer Ki-67 expression. The models were internally and externally validated, and the AUC of each model was calculated. Calibration and decision curves were used to evaluate the clinical practicality of nomograms.ResultsThree models for predicting rectal cancer Ki-67 expression were constructed. The AUC and Delong test results revealed that the combined model had better prediction performance than other models in three chohrts. A decision curve analysis revealed that the nomogram based on the combined model had relatively good clinical performance, which can be an intuitive prediction tool for clinicians.ConclusionThe multiparametric MRI radiomics model can provide a noninvasive and accurate auxiliary tool for preoperative evaluation of Ki-67 expression in patients with rectal cancer and can support clinical decision-making.

Machine Learning Models in Prediction of Treatment Response After Chemoembolization with MRI Clinicoradiomics Features.

To evaluate machine learning models, created with radiomics and clinicoradiomics features, ability to predict local response after TACE. 188 treatment-naïve patients (150 responders, 38 non-responders) with HCC who underwent TACE were included in this retrospective study. Laboratory, clinical and procedural information were recorded. Local response was evaluated by European Association for the Study of the Liver criteria at 3-months. Radiomics features were extracted from pretreatment pre-contrast enhanced T1 (T1WI) and late arterial-phase contrast-enhanced T1 (CE-T1) MRI images. After data augmentation, data were split into training and test sets (70/30). Intra-class correlations, Pearson's correlation coefficients were analyzed and followed by a sequential-feature-selection (SFS) algorithm for feature selection. Support-vector-machine (SVM) models were trained with radiomics and clinicoradiomics features of T1WI, CE-T1 and the combination of both datasets, respectively. Performance metrics were calculated with the test sets. Models' performances were compared with Delong's test. 1128 features were extracted. In feature selection, SFS algorithm selected 18, 12, 24 and 8 features in T1WI, CE-T1, combined datasets and clinical features, respectively. The SVM models area-under-curve was 0.86 and 0.88 in T1WI; 0.76, 0.71 in CE-T1 and 0.82, 0.91 in the combined dataset, with and without clinical features, respectively. The only significant change was observed after inclusion of clinical features in the combined dataset (p = 0.001). Higher WBC and neutrophil levels were significantly associated with lower treatment response in univariant analysis (p = 0.02, for both). Machine learning models created with clinical and MRI radiomics features, may have promise in predicting local response after TACE. Level 4, Case-control study.

Exploring the association of glioma tumor residuals from incongruent [18F]FET PET/MR imaging with tumor proliferation using a multiparametric MRI radiomics nomogram.

The study aimed to using multiparametric MRI radiomics to predict glioma tumor residuals (TRFET over MR) derived from incongruent [18F]fluoroethyl-L-tyrosine ([18F]FET) PET/MR imaging. One hundred ten patients with gliomas who underwent [18F]FET PET/MR scanning were retrospectively analyzed. The TRFET over MR was identified by the discrepancy-PET that the extent of resection (EOR) based on MRI subtracted the biological tumor volume on PET images. The MRI parameters and radiomics features were extracted based on EOR and selected by the least absolute shrinkage and selection operator to construct radiomics score (Rad-score). The correlation network analysis of all features was analyzed by Spearman's correlation tests. The methods for evaluating the clinical usefulness consisted of the receiver operating characteristic curve, the calibration curve, and decision curve analysis. The Rad-score of the patients with the TRFET over MR was significantly higher than those with the non TRFET over MR (p < 0.001). The Rad-score was significantly correlated with the discrepancy-PET (r = 0.72, p < 0.001), Ki-67 level (r = 0.76, p < 0.001), and epidermal growth factor receptor (EGFR) of gliomas (r = 0.75, p < 0.001), respectively. Moreover, there was a difference of the correlation network analysis between the TRPET over MR group and non TRFET over MR group. The nomogram combing Rad-score and clinical features had the greatest performance in predicting TRFET over MR (AUC = 0.90/0.87, training/testing). There was a significant difference in prognosis (median OS, 17m vs. 43m) between patients with TRFET over MR and non TRFET over MR based on nomogram prediction (p < 0.001). The nomogram based on MRI radiomics would predict gliomas tumor residuals caused by the absence of 18F-PET PET examination and adjust EOR to improve prognosis.

Establishment of a Prediction Model Based on Preoperative MRI Radiomics for Diffuse Astrocytic Glioma, IDH-Wildtype, with Molecular Features of Glioblastoma.

In 2021, the WHO central nervous system (CNS) tumor classification criteria added the diagnosis of diffuse astrocytic glioma, IDH wild-type, with molecular features of glioblastoma, WHO grade 4 (DAG-G). DAG-G may exhibit the aggressiveness and malignancy of glioblastoma (GBM) despite the lower histological grade, and thus a precise preoperative diagnosis can help neurosurgeons develop more refined individualized treatment plans. This study aimed to establish a predictive model for the non-invasive identification of DAG-G based on preoperative MRI radiomics. Patients with pathologically confirmed glioma in Huashan Hospital, Fudan University, between September 2019 and July 2021 were retrospectively analyzed. Furthermore, two external validation datasets from Wuhan Union Hospital and Xuzhou Cancer Hospital were also utilized to verify the reliability and accuracy of the prediction model. Two regions of interest (ROI) were delineated on the preoperative MRI images of the patients using the semi-automatic tool ITK-SNAP (version 4.0.0), which were named the maximum anomaly region (ROI1) and the tumor region (ROI2), and Pyradiomics 3.0 was applied for feature extraction. Feature selection was performed using a least absolute shrinkage and selection operator (LASSO) filter and a Spearman correlation coefficient. Six classifiers, including Gauss naive Bayes (GNB), K-nearest neighbors (KNN), Random forest (RF), Adaptive boosting (AB), and Support vector machine (SVM) with linear kernel and multilayer perceptron (MLP), were used to build the prediction models, and the prediction performance of the six classifiers was evaluated by fivefold cross-validation. Moreover, the performance of prediction models was evaluated using area under the curve (AUC), precision (PRE), and other metrics. According to the inclusion and exclusion criteria, 172 patients with grade 2-3 astrocytoma were finally included in the study, and a total of 44 patients met the diagnosis of DAG-G. In the prediction task of DAG-G, the average AUC of GNB classifier was 0.74 ± 0.07, that of KNN classifier was 0.89 ± 0.04, that of RF classifier was 0.96 ± 0.03, that of AB classifier was 0.97 ± 0.02, that of SVM classifier was 0.88 ± 0.05, and that of MLP classifier was 0.91 ± 0.03, among which, AB classifier achieved the best prediction performance. In addition, the AB classifier achieved AUCs of 0.91 and 0.89 in two external validation datasets obtained from Wuhan Union Hospital and Xuzhou Cancer Hospital, respectively. The prediction model constructed based on preoperative MRI radiomics established in this study can basically realize the prospective, non-invasive, and accurate diagnosis of DAG-G, which is of great significance to help further optimize treatment plans for such patients, including expanding the extent of surgery and actively administering radiotherapy, targeted therapy, or other treatments after surgery, to fundamentally maximize the prognosis of patients.

Assessment of Lymphovascular Invasion in Breast Cancer Using a Combined MRI Morphological Features, Radiomics, and Deep Learning Approach Based on Dynamic Contrast-Enhanced MRI.

Assessment of lymphovascular invasion (LVI) in breast cancer (BC) primarily relies on preoperative needle biopsy. There is an urgent need to develop a non-invasive assessment method. To develop an effective model to assess the LVI status in patients with BC using magnetic resonance imaging morphological features (MRI-MF), Radiomics, and deep learning (DL) approaches based on dynamic contrast-enhanced MRI (DCE-MRI). Cross-sectional retrospective cohort study. The study included 206 BC patients, with 136 in the training set [97 LVI(-) and 39 LVI(+) cases; median age: 51.5 years] and 70 in the test set [52 LVI(-) and 18 LVI(+) cases; median age: 48 years]. 1.5 T/T1-weighted images, fat-suppressed T2-weighted images, diffusion-weighted imaging (DWI), and DCE-MRI. The MRI-MF model was developed with conventional MR features using logistic analyses. The Radiomic feature extraction process involved collecting data from categorized DCE-MRI datasets, specifically the first and second post-contrast images (A1 and A2). Next, a DL model was implemented to determine LVI. Finally, we established a joint diagnosis model by combining the MRI-MF, Radiomics, and DL approaches. Diagnostic performance was compared using receiver operating characteristic curve analysis, confusion matrix, and decision curve analysis. Rim sign and peritumoral edema features were used to develop the MRI-MF model, while six Radiomics signature from the A1 and A2 images were used for the Radiomics model. The joint model (MRI-MF + Radiomics + DL models) achieved the highest accuracy (area under the curve [AUC] = 0.857), being significantly superior to the MRI-MF (AUC = 0.724), Radiomics (AUC = 0.736), or DL (AUC = 0.740) model. Furthermore, it also outperformed the pairwise combination models: Radiomics + MRI-MF (AUC = 0.796), DL + MRI-MF (AUC = 0.796), or DL + Radiomics (AUC = 0.826). The joint model incorporating MRI-MF, Radiomics, and DL approaches can effectively determine the LVI status in patients with BC before surgery. 4 TECHNICAL EFFICACY: Stage 2.

Effects of MRI radiomics combined with clinical data in evaluating lymph node metastasis in mrT1-3a staging rectal cancer.

To investigate the value of a clinical-MRI radiomics model based on clinical characteristics and T2-weighted imaging (T2WI) for preoperatively evaluating lymph node (LN) metastasis in patients with MRI-predicted low tumor (T) staging rectal cancer (mrT1, mrT2, and mrT3a with extramural spread ≤ 5mm). This retrospective study enrolled 303 patients with low T-staging rectal cancer (training cohort, n = 213, testing cohort n = 90). A total of 960 radiomics features were extracted from T2WI. Minimum redundancy and maximum relevance (mRMR) and support vector machine were performed to select the best performed radiomics features for predicting LN metastasis. Multivariate logistic regression analysis was then used to construct the clinical and clinical-radiomics combined models. The model performance for predicting LN metastasis was assessed by receiver operator characteristic curve (ROC) and clinical utility implementing a nomogram and decision curve analysis (DCA). The predictive performance for LN metastasis was also compared between the combined model and human readers (2 seniors). Fourteen radiomics features and 2 clinical characteristics were selected for predicting LN metastasis. In the testing cohort, a higher positive predictive value of 75.9% for the combined model was achieved than those of the clinical model (44.8%) and two readers (reader 1: 54.9%, reader 2: 56.3%) in identifying LN metastasis. The interobserver agreement between 2 readers was moderate with a kappa value of 0.416.A clinical-radiomics nomogram and decision curve analysis demonstrated that the combined model was clinically useful. T2WI-based radiomics combined with clinical data could improve the efficacy in noninvasively evaluating LN metastasis for the low T-staging rectal cancer and aid in tailoring treatment strategies.

MRI radiomics for predicting poor disease-free survival in muscle invasive bladder cancer: the results of the retrospective cohort study.

To develop an MRI radiomic nomogram capable of identifying muscle invasive bladder cancer (MIBC) patients with high-risk molecular characteristics related to poor 2-year disease-free survival (DFS). We performed a retrospective analysis of DNA sequencing data, prognostic information, and radiomics features from 91 MIBC patients at stages T2-T4aN0M0 without history of immunotherapy. To identify risk stratification, we employed Cox regression based on TP53 mutation status and tumor mutational burden (TMB) level. Radiomics signatures were selected using the least absolute shrinkage and selection operator (LASSO) to construct a nomogram based on logistic regression for predicting the stratification in the training cohort. The predictive performance of the nomogram was assessed in the testing cohort using receiver operator curve (ROC), Hosmer-Lemeshow (HL) test, clinical impact curve (CIC), and decision curve analysis (DCA). Among 91 participants, the mean TMB value was 3.3 mut/Mb, with 60 participants having TP53 mutations. Patients with TP53 mutations and a below-average TMB value were identified as high risk and had a significantly poor 2-year DFS (hazard ratio = 4.36, 95% CI 1.82-10.44, P < 0.001). LASSO identified five radiomics signatures that correlated with the risk stratification. In the testing cohort, the nomogram achieved an area under the ROC curve of 0.909 (95% CI 0.789-0.991) and an accuracy of 0.889 (95% CI 0.708-0.977). The molecular risk stratification based on TP53 mutation status combined with TMB level is strongly associated with DFS in MIBC. Radiomics signatures can effectively predict this stratification and provide valuable information to clinical decision-making.

Noninvasive prediction of IDH mutation status in gliomas using preoperative multiparametric MRI radiomics nomogram: A mutlicenter study

PurposeTo establish and validate a radiomics nomogram for preoperative prediction of isocitrate dehydrogenase (IDH) mutation status of gliomas in a multicenter setting. Methods414 gliomas patients were collected (306 from local institution and 108 from TCGA). 851 radiomics features were extracted from contrast-enhanced T1-weighted (CE-T1W) and fluid attenuated inversion recovery (FLAIR) sequence, respectively. The features were refined using least absolute shrinkage and selection operator (LASSO) regression combing 10-fold cross-validation. The optimal radiomics features with age and sex were processed by multivariate logistic regression analysis to construct a prediction model, which was developed in the training dataset and assessed in the test and validation dataset. Receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis were applied in the test and external validation datasets to evaluate the performance of the prediction model. ResultsTen robust radiomics features were selected from the 1702 features (four CE-T1W features and six FLAIR features). A nomogram was plotted to represent the prediction model. The accuracy and AUC of the radiomics nomogram achieved 86.96% and 0.891(0.809–0.947) in the test dataset and 84.26% and 0.881(0.805–0.936) in the external validation dataset (all p < 0.05). The positive predictive value (PPV) and negative predictive value (NPV) were 83.72% and 87.75% in the test dataset and 87.81% and 82.09% in the external validation dataset. ConclusionIDH genotypes of gliomas can be identified by preoperative multiparametric MRI radiomics nomogram and might be clinically meaningful for treatment strategy and prognosis stratification of gliomas.

A brain tumor computer-aided diagnosis method with automatic lesion segmentation and ensemble decision strategy.

Gliomas and brain metastases (Mets) are the most common brain malignancies. The treatment strategy and clinical prognosis of patients are different, requiring accurate diagnosis of tumor types. However, the traditional radiomics diagnostic pipeline requires manual annotation and lacks integrated methods for segmentation and classification. To improve the diagnosis process, a gliomas and Mets computer-aided diagnosis method with automatic lesion segmentation and ensemble decision strategy on multi-center datasets was proposed. Overall, 1,022 high-grade gliomas and 775 Mets patients' preoperative MR images were adopted in the study, including contrast-enhanced T1-weighted (T1-CE) and T2-fluid attenuated inversion recovery (T2-flair) sequences from three hospitals. Two segmentation models trained on the gliomas and Mets datasets, respectively, were used to automatically segment tumors. Multiple radiomics features were extracted after automatic segmentation. Several machine learning classifiers were used to measure the impact of feature selection methods. A weight soft voting (RSV) model and ensemble decision strategy based on prior knowledge (EDPK) were introduced in the radiomics pipeline. Accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC) were used to evaluate the classification performance. The proposed pipeline improved the diagnosis of gliomas and Mets with ACC reaching 0.8950 and AUC reaching 0.9585 after automatic lesion segmentation, which was higher than those of the traditional radiomics pipeline (ACC:0.8850, AUC:0.9450). The proposed model accurately classified gliomas and Mets patients using MRI radiomics. The novel pipeline showed great potential in diagnosing gliomas and Mets with high generalizability and interpretability.

Predictive Value of Pre-Treatment MRI Radiomics for Distant Brain Metastases Following Stereotactic Radiosurgery/Radiotherapy

Local intracranial therapy for brain metastases (BM) has taken on particular importance as survival among metastatic patients improves. However, the development of distant BMs (DBMs) outside the treated area remains a stubborn problem for which canonical clinical features (age, histology, ECOG PS) have limited predictive capability. In this study, we hypothesized that MRI-based "radiomic" features (sub-visual cues extracted from diagnostic images) can accurately predict the time-to-DBM development (TTDD) on a retrospectively curated dataset of patients treated with stereotactic radiosurgery/radiotherapy (SRS/SRT). We queried our treatment planning system for patients treated with brain SRS/SRT between 2014 and 2021, and curated the incidence/timing of DBMs manually. Pre-RT MRI sequences (T1 pre, T1 post, T2, and FLAIR) and planning data were obtained for each patient. MRI and CT simulations were co-registered using affine transformations, and regions of interest (ROIs) were identified based on contoured structures (GTV) and discrete isodose ranges (0-25%, 25-50%, 50-75%, 75%+). Radiomic features were extracted from these ROIs, and clinical features (ECOG PS, tumor volume, age) were recorded for baseline comparison. Features were selected using Wald test scores from univariate Cox proportional hazard (CPH) models. Multivariate CPH models were then trained to predict TTDD using combinations of selected features. Predictive capability was evaluated using concordance index (c-index) values. A radiomic risk score (RRS) was created to discriminate patients with low and high-risk for DBMs, and evaluated using a log-rank test. A total of 105 patients were selected with a median follow up of 356 days. 53 patients developed DBMs (median time 118 days). Radiomic CPH models achieved a c-index of 0.63 compared to clinical baseline of 0.49. The combination of radiomic and clinical features achieved the highest c-index of 0.69. Overall, radiomic features with and without clinical features were able to stratify patients into low and high-risk groups with statistically significant differences in TTDD development (see Table 1). Clinical features alone were not significant. The most predictive radiomic features were identified within the T1 pre-contrast MRI from the 50-75% isodose regions, followed by T2 FLAIR/GTV and T2/GTV combinations. Radiomic features from routine MR scans were more predictive of TTDD than baseline clinical features. The contribution from the 50-75% isodose region suggests importance within the peritumoral environment in addition to the tumor itself.

MRI radiomics for the preoperative evaluation of lymphovascular invasion in breast cancer: A meta-analysis

PurposeTo evaluate the ability of preoperative MRI-based radiomic features in predicting lymphovascular invasion (LVI) in patients with breast cancer. MethodsPubMed, Embase, Web of Science, Cochrane Library databases, and four Chinese databases were searched to identify relevant studies published up until June 15, 2023. Two reviewers screened all papers independently for eligibility. We included diagnostic accuracy studies that used radiomics-MRI for LVI in patients with breast cancer, using histopathology as the reference standard. Quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and Radiomics Quality Score. Overall diagnostic odds ratio (DOR), sensitivity, specificity and area under the curve (AUC) were calculated to assess the prediction efficacy of MRI-based radiomic features in patients with breast cancer. Spearman’s correlation coefficient was calculated and subgroup analysis performed to investigate causes of heterogeneity. ResultsEight studies comprising 1685 female patients were included. The pooled DOR, sensitivity, specificity, and AUC of radiomics in detecting LVI were 23 [confidence interval (CI) 16,32], 0.89(0.86,0.92), 0.82 (0.78,0.86), and 0.83(0.78,0.87), respectively. The meta-analysis showed significant heterogeneity among the included studies. No threshold effect was detected. Subgroup analysis showed that more than 200 participants, radiomics with clinical factors, semiautomatic segmentation method and peritumoral or intra- and peritumoral model [DOR: 28(18,42), 26(19,37), 34(16,70), 40(10,156), respectively] could improve diagnostic performance compared with less than 200 participants, only radiomics, manual segmentation method, and tumor model [DOR: 16(7,37), 21(6,73), 20(12,32), 21(13,32), respectively], but 3.0 T MR and multiple sequences approach [DOR: 27(15,49),17(8,35)] couldn’t improve diagnostic performance compared with 1.5 T and DCE radiomic features [DOR:27(7,99),25(17,37)]. ConclusionOur meta-analysis showed that preoperative MRI-based radiomic features performs well in predicting LVI in patients with breast cancer. This noninvasive and convenient tool may be used to facilitate preoperative identification of LVI in breast cancer.

Using MRI Radiomics to Predict the Efficacy of Immunotherapy for Brain Metastasis in Patients with Non-small Cell Lung Cancer

Despite immune checkpoint inhibitors (ICIs) has been proven effective in patients with advanced NSCLC, controversial therapeutic responses are seen in brain metastatic lesions. There are no robust biomarkers that predict benefit from this regimen. We evaluated the utility of novel imaging biomarkers (radiomics) to distinguish NSCLC patients with brain metastases (BMs) who will benefit from ICIs from those likely to progress despite therapy. One-hundred and seventy-four NSCLC Patients with BMs treated with ICIs from June 2019 - June 2022 were identified. We collected patient clinical outcomes and pre-treatment MRI images. Images were split into training and test sets. Brain metastatic lesions were contoured on ITK-SNAP software and 3748 radiomic features capturing both intra- and peritumoral texture patterns were extracted. The primary endpoint of this study was intracranial progression-free survival (iPFS) and the secondary objective were progression-free survival (PFS) and overall survival (OS). We used the least absolute shrinkage and selection operator (LASSO) Cox regression model to build the radiomic signature for ORR. Based on the support vector machine (SVM) model, we construct the Clinical-radiomics nomogram (CRN). Multivariable Cox regression analysis were performed to evaluate the effect of each factor on iPFS. We performed Kaplan-Meier survival analysis and log-rank tests to assess prognostic value of the features. We identified 174 patients who fit our criteria with available MRI images. 122 patients treated in our center were divided into a training set and 52 patients treated in another center were divided into a test set. The intratumoral radiomic signatures(IRS), peritumoral radiomic signatures(PRS) and CRN showed favorable predictive effects for ORR with the area under the receiver operating curve (AUC) of (IRS: 0.845 (95% CI: 0.776-0.914); PRS: 0.799 (95% CI:0.720-0.879); CRN:0.907 (95% CI: 0.855-0.959))in the training set and (IRS: 0.809 (95% CI: 0.693-0.926); PRS: 0.749 (95% CI:0.616-0.883); CRN:0.888(95% CI:0.798-0.979)) in the test set. Kaplan-Meier analyses showed a significantly longer iPFS in the high-CRN group versus the low-CRN group (P < 0.001). The CRNs were also found significantly associated with PFS (P < 0.001), but not OS. Radiomic biomarkers from pre-treatment MRI images in NSCLC patients with BMs were predictive of iPFS to ICIs. Pre-treatment radiomics may allow early prediction of benefit and expedite more aggressive treatment for high-risk patients. Additional validation of these imaging biomarkers is warranted.

Biopsy-free AI-aided precision MRI assessment in prediction of prostate cancer biochemical recurrence.

To investigate the predictive ability of high-throughput MRI with deep survival networks for biochemical recurrence (BCR) of prostate cancer (PCa) after prostatectomy. Clinical-MRI and histopathologic data of 579 (train/test, 463/116) PCa patients were retrospectively collected. The deep survival network (iBCR-Net) is based on stepwise processing operations, which first built an MRI radiomics signature (RadS) for BCR, and predicted the T3 stage and lymph node metastasis (LN+) of tumour using two predefined AI models. Subsequently, clinical, imaging and histopathological variables were integrated into iBCR-Net for BCR prediction. RadS, derived from 2554 MRI features, was identified as an independent predictor of BCR. Two predefined AI models achieved an accuracy of 82.6% and 78.4% in staging T3 and LN+. The iBCR-Net, when expressed as a presurgical model by integrating RadS, AI-diagnosed T3 stage and PSA, can match a state-of-the-art histopathological model (C-index, 0.81 to 0.83 vs 0.79 to 0.81, p > 0.05); and has maximally 5.16-fold, 12.8-fold, and 2.09-fold (p < 0.05) benefit to conventional D'Amico score, the Cancer of the Prostate Risk Assessment (CAPRA) score and the CAPRA Postsurgical score. AI-aided iBCR-Net using high-throughput MRI can predict PCa BCR accurately and thus may provide an alternative to the conventional method for PCa risk stratification.

Adjuvant chemotherapy or no adjuvant chemotherapy? A prediction model for the risk stratification of recurrence or metastasis of nasopharyngeal carcinoma combining MRI radiomics with clinical factors.

Dose adjuvant chemotherapy (AC) should be offered in nasopharyngeal carcinoma (NPC) patients? Different guidelines provided the different recommendations. In this retrospective study, a total of 140 patients were enrolled and followed for 3 years, with 24 clinical features being collected. The imaging features on the enhanced-MRI sequence were extracted by using PyRadiomics platform. The pearson correlation coefficient and the random forest was used to filter the features associated with recurrence or metastasis. A clinical-radiomics model (CRM) was constructed by the Cox multivariable analysis in training cohort, and was validated in validation cohort. All patients were divided into high- and low-risk groups through the median Rad-score of the model. The Kaplan-Meier survival curves were used to compare the 3-year recurrence or metastasis free rate (RMFR) of patients with or without AC in high- and low-groups. In total, 960 imaging features were extracted. A CRM was constructed from nine features (seven imaging features and two clinical factors). In the training cohort, the area under curve (AUC) of CRM for 3-year RMFR was 0.872 (P <0.001), and the sensitivity and specificity were 0.935 and 0.672, respectively; In the validation cohort, the AUC was 0.864 (P <0.001), and the sensitivity and specificity were 1.00 and 0.75, respectively. Kaplan-Meier curve showed that the 3-year RMFR and 3-year cancer specific survival (CSS) rate in the high-risk group were significantly lower than those in the low-risk group (P <0.001). In the high-risk group, patients who received AC had greater 3-year RMFR than those who did not receive AC (78.6% vs. 48.1%) (p = 0.03). Considering increasing RMFR, a prediction model for NPC based on two clinical factors and seven imaging features suggested the AC needs to be added to patients in the high-risk group and not in the low-risk group.

Nomograms Based on MRI Radiomics for Differential Diagnosis and Predicting BRAFV600E Expression in Pleomorphic Xanthoastrocytoma and Ganglioglioma

This study was designed to investigate the value of nomograms based on MRI radiomics and clinical semantic features in identifying pleomorphic xanthoastrocytoma (PXA) and ganglioglioma (GG) as well as predicting BRAFV600E expression. This study included 265 patients histologically diagnosed with PXA (n=113) and GG (n=152). T1WI, T2WI, and CET1 sequences were utilized to extract radiomics features. Univariate analysis, Spearman correlation analysis, and the least absolute shrinkage and selection operator were used for dimensionality reduction and feature selection. Following this, logistic regression was utilized to establish the radiomics model. Univariate and multivariate analyses of clinical semantic features were applied, and clinical models were constructed. The nomograms were established by merging radiomics and clinical features. Furthermore, ROC curve analysis was used for examining the model performance, whereas the decision curve analysis (DCA) examined the clinical utility of the nomograms. Nomograms achieved the best predictive efficacy compared to clinical and radiomics models alone. Concerning the differentiation between PXA and GG, the area under the curve (AUC) values of the nomogram were 0.879 (0.828-0.930) and 0.887 (0.805-0.969) for the training and testing cohorts, respectively. For predicting BRAFV600E expression, the AUC values of the nomogram were 0.873 (0.811-0.936) and 0.851 (0.740-0.963) for the training and testing cohorts, respectively. DCA confirmed the clinical utility of the nomograms. Nomograms based on radiomics and clinical semantic features were noninvasive tools for differential diagnosis of PXA and GG and predicting BRAFV600E expression, which may be helpful for assessing patient prognosis and developing individualized treatment strategies.

MRI radiomics for hamstring strain injury identification and return to sport classification: a pilot study.

To determine if MRI-based radiomics from hamstring muscles are related to injury and if the features could be used to perform a time to return to sport (RTS) classification. We hypothesize that radiomics from hamstring muscles, especially T2-weighted and diffusion tensor imaging-based features, are related to injury and can be used for RTS classification. MRI data from 32 athletes at the University of Wisconsin-Madison that sustained a hamstring strain injury were collected. Diffusion tensor imaging and T1- and T2-weighted images were processed, and diffusion maps were calculated. Radiomics features were extracted from the four hamstring muscles in each limb and for each MRI modality, individually. Feature selection was performed and multiple support vector classifiers were cross-validated to differentiate between involved and uninvolved limbs and perform binary (≤ or > 25days) and multiclass (< 14 vs. 14-42 vs. > 42days) classification of RTS. The combination of radiomics features from all diffusion tensor imaging and T2-weighted images provided the most accurate differentiation between involved and uninvolved limbs (AUC ≈ 0.84 ± 0.16). For the binary RTS classification, the combination of all extracted radiomics offered the most accurate classification (AUC ≈ 0.95 ± 0.15). While for the multiclass RTS classification, the combination of features from all the diffusion tensor imaging maps provided the most accurate classification (weighted one vs. rest AUC ≈ 0.81 ± 0.16). This pilot study demonstrated that radiomics features from hamstring muscles are related to injury and have the potential to predict RTS.

Development of a nomogram based on radiomics and semantic features for predicting chromosome 7 gain/chromosome 10 loss in IDH wild-type histologically low-grade gliomas.

To predict chromosome 7 gain and chromosome 10 loss (+7/-10) in IDH wild-type (IDH-wt) histologically low-grade gliomas (LGG) by machine learning models based on MRI radiomics and semantic features. A total of 122 patients diagnosed as IDH-wt histologically LGG were retrospectively included in this study. The patients were randomly divided into a training group and a test group in a ratio of 7:3. The radiomics features were extracted from axial T1WI, T2WI, FLAIR and CET1 sequences, respectively. The distance correlation (DC) and least absolute shrinkage and selection operator (LASSO) were used to select the radiomics signatures. Three machine learning algorithms including neural network (NN), support vector machine (SVM), and linear discriminant analysis (LDA) were used to construct radiomics models. In addition, a nomogram was developed by combining the optimal radiomics signature with clinical risk factors, and the potential clinical utility of the nomogram was evaluated using decision curve analysis. The LDA+DC model was identified as the optimal classifier among the six radiomics models. Necrosis was determined as a risk factor for +7/-10 in IDH-wt histologically LGG. The nomogram achieved the best performance, with an AUC of 0.854 and an accuracy of 0.778 in the independent test group. The decision curve of the nomogram confirmed its clinical usefulness in a wide range of thresholds. The nomogram combining radiomics and semantic features can predict the +7/-10 status effectively, which may contribute to the risk stratification and individualized treatment planning of patients with IDH-wt histologically LGG.

Radiomic features from multiparametric magnetic resonance imaging predict molecular subgroups of pediatric low-grade gliomas

BackgroundWe aimed to develop machine learning models for prediction of molecular subgroups (low-risk group and intermediate/high-risk group) and molecular marker (KIAA1549-BRAF fusion) of pediatric low-grade gliomas (PLGGs) based on radiomic features extracted from multiparametric MRI.Methods61 patients with PLGGs were included in this retrospective study, which were divided into a training set and an internal validation set at a ratio of 2:1 based on the molecular subgroups or the molecular marker. The patients were classified into low-risk and intermediate/high-risk groups, BRAF fusion positive and negative groups, respectively. We extracted 5929 radiomic features from multiparametric MRI. Thereafter, we removed redundant features, trained random forest models on the training set for predicting the molecular subgroups or the molecular marker, and validated their performance on the internal validation set. The performance of the prediction model was verified by 3-fold cross-validation.ResultsWe constructed the classification model differentiating low-risk PLGGs from intermediate/high-risk PLGGs using 4 relevant features, with an AUC of 0.833 and an accuracy of 76.2% in the internal validation set. In the prediction model for predicting KIAA1549-BRAF fusion using 4 relevant features, an AUC of 0.818 and an accuracy of 81.0% were achieved in the internal validation set.ConclusionsThe current study demonstrates that MRI radiomics is able to predict molecular subgroups of PLGGs and KIAA1549-BRAF fusion with satisfying sensitivity.Trial registrationThis study was retrospectively registered at clinicaltrials.gov (NCT04217018).

T2-Weighted MRI Radiomic Features Predict Prostate Cancer Presence and Eventual Biochemical Recurrence.

Prostate cancer (PCa) is the most diagnosed non-cutaneous cancer in men. Despite therapies such as radical prostatectomy, which is considered curative, distant metastases may form, resulting in biochemical recurrence (BCR). This study used radiomic features calculated from multi-parametric magnetic resonance imaging (MP-MRI) to evaluate their ability to predict BCR and PCa presence. Data from a total of 279 patients, of which 46 experienced BCR, undergoing MP-MRI prior to surgery were assessed for this study. After surgery, the prostate was sectioned using patient-specific 3D-printed slicing jigs modeled using the T2-weighted imaging (T2WI). Sectioned tissue was stained, digitized, and annotated by a GU-fellowship trained pathologist for cancer presence. Digitized slides and annotations were co-registered to the T2WI and radiomic features were calculated across the whole prostate and cancerous lesions. A tree regression model was fitted to assess the ability of radiomic features to predict BCR, and a tree classification model was fitted with the same radiomic features to classify regions of cancer. We found that 10 radiomic features predicted eventual BCR with an AUC of 0.97 and classified cancer at an accuracy of 89.9%. This study showcases the application of a radiomic feature-based tool to screen for the presence of prostate cancer and assess patient prognosis, as determined by biochemical recurrence.

Achieving imaging and computational reproducibility on multiparametric MRI radiomics features in brain tumor diagnosis: phantom and clinical validation.

The Image Biomarker Standardization Initiative has helped improve the computational reproducibility of MRI radiomics features. Nonetheless, the MRI sequences and features with high imaging reproducibility are yet to be established. To determine reproducible multiparametric MRI radiomics features across test-retest, multi-scanner, and computational reproducibility comparisons, and to evaluate their clinical value in brain tumor diagnosis. To assess reproducibility, T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and diffusion-weighted imaging (DWI) were acquired from three 3-T MRI scanners using standardized phantom, and radiomics features were extracted using two computational algorithms. Reproducible radiomics features were selected when the concordance correlation coefficient value above 0.9 across multiple sessions, scanners, and computational algorithms. Random forest classifiers were trained with reproducible features (n = 117) and validated in a clinical cohort (n = 50) to evaluate whether features with high reproducibility improved the differentiation of glioblastoma from primary central nervous system lymphomas (PCNSLs). Radiomics features from T2WI demonstrated higher repeatability (65-94%) than those from DWI (38-48%) or T1WI (2-92%). Across test-retest, multi-scanner, and computational comparisons, T2WI provided 41 reproducible features, DWI provided six, and T1WI provided two. The performance of the classification model with reproducible features was higher than that using non-reproducible features in both training set (AUC, 0.916 vs. 0.877) and validation set (AUC, 0.957 vs. 0.869). Radiomics features with high reproducibility across multiple sessions, scanners, and computational algorithms were identified, and they showed higher diagnostic performance than non-reproducible radiomics features in the differentiation of glioblastoma from PCNSL. By identifying the radiomics features showing higher multi-machine reproducibility, our results also demonstrated higher radiomics diagnostic performance in the differentiation of glioblastoma from PCNSL, paving the way for further research designs and clinical application in neuro-oncology. • Highly reproducible radiomics features across multiple sessions, scanners, and computational algorithms were identified using phantom and applied to clinical diagnosis. • Radiomics features from T2-weighted imaging were more reproducible than those from T1-weighted and diffusion-weighted imaging. • Radiomics features with good reproducibility had better diagnostic performance for brain tumors than features with poor reproducibility.