Preoperative Prediction of Meningioma Subtype by Constructing a Clinical-Radiomics Model Nomogram Based on Magnetic Resonance Imaging.

OBJECTIVE: We sought to investigate the value of a clinical-radiomics model based on magnetic resonance imaging in differentiating fibroblastic meningiomas from non-fibroblastic meningiomas. METHODS: Clinical, imaging, and postoperative pathologic data of 423 patients (128 fibroblastic meningiomas and 295 non-fibroblastic meningiomas) were randomly categorized into training (n = 296) and validation (n = 127) groups at a 7:3 ratio. The Selectpercentile and LASSO were used to selected the highly correlated features from 3376 radiomics features. Different classifiers were used to train and verify the model. The receiver operating characteristic curves, accuracy (ACC), sensitivity (SEN), and specificity (SPE) were drawn to evaluate the performance. The optimal radiomics model was selected. Calibration curves and decision curve analysis were used to verify the clinical utility and consistency of the nomogram constructed from the radiomics features and clinical factors. RESULTS: Thirteen radiomics features were selected from contrast-enhanced T1-weighted imaging and T2-weighted imaging after dimensionality reduction. The prediction performance of random forest radiomics model is slightly lower than that of the clinical-radiomics model. The area under the curve, SEN, SPE, and ACC of the clinical-radiomics model training set were 0.836 (95% confidence interval, 0.795-0.878), 0.922, 0.583, and 0.686, respectively. The area under the curve, SEN, SPE, and ACC of the validation set were 0.756 (95% confidence interval, 0.660-0.846), 0.816, 0.596, and 0.661, respectively. CONCLUSIONS: The diagnostic efficacy of the clinical-radiomics model of fibroblastic meningioma and non-fibroblastic meningioma was better than that of the radiomics prediction model alone and can be used as a potential tool for clinical surgical planning and evaluation of patient prognosis.

Magnetic Resonance Deep Learning Radiomic Model Based on Distinct Metastatic Vascular Patterns for Evaluating Recurrence-Free Survival in Hepatocellular Carcinoma.

BACKGROUND: The metastatic vascular patterns of hepatocellular carcinoma (HCC) are mainly microvascular invasion (MVI) and vessels encapsulating tumor clusters (VETC). However, most existing VETC-related radiological studies still focus on the prediction of VETC status. PURPOSE: This study aimed to build and compare VETC-MVI related models (clinical, radiomics, and deep learning) associated with recurrence-free survival of HCC patients. STUDY TYPE: Retrospective. POPULATION: 398 HCC patients (349 male, 49 female; median age 51.7 years, and age range: 22-80 years) who underwent resection from five hospitals in China. The patients were randomly divided into training cohort (n = 358) and test cohort (n = 40). FIELD STRENGTH/SEQUENCE: 3-T, pre-contrast T1-weighted imaging spoiled gradient recalled echo (T1WI SPGR), T2-weighted imaging fast spin echo (T2WI FSE), and contrast enhanced arterial phase (AP), delay phase (DP). ASSESSMENT: Two radiologists performed the segmentation of HCC on T1WI, T2WI, AP, and DP images, from which radiomic features were extracted. The RFS related clinical characteristics (VETC, MVI, Barcelona stage, tumor maximum diameter, and alpha fetoprotein) and radiomic features were used to build the clinical model, clinical-radiomic (CR) nomogram, deep learning model. The follow-up process was done 1 month after resection, and every 3 months subsequently. The RFS was defined as the date of resection to the date of recurrence confirmed by radiology or the last follow-up. Patients were followed up until December 31, 2022. STATISTICAL TESTS: Univariate COX regression, least absolute shrinkage and selection operator (LASSO), Kaplan-Meier curves, log-rank test, C-index, and area under the curve (AUC). P < 0.05 was considered statistically significant. RESULTS: The C-index of deep learning model achieved 0.830 in test cohort compared with CR nomogram (0.731), radiomic signature (0.707), and clinical model (0.702). The average RFS of the overall patients was 26.77 months (range 1-80 months). DATA CONCLUSION: MR deep learning model based on VETC and MVI provides a potential tool for survival assessment. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 3.

Baseline whole-lung CT features deriving from deep learning and radiomics: prediction of benign and malignant pulmonary ground-glass nodules.

Objective: To develop and validate the model for predicting benign and malignant ground-glass nodules (GGNs) based on the whole-lung baseline CT features deriving from deep learning and radiomics. Methods: This retrospective study included 385 GGNs from 3 hospitals, confirmed by pathology. We used 239 GGNs from Hospital 1 as the training and internal validation set; 115 and 31 GGNs from Hospital 2 and Hospital 3 as the external test sets 1 and 2, respectively. An additional 32 stable GGNs from Hospital 3 with more than five years of follow-up were used as the external test set 3. We evaluated clinical and morphological features of GGNs at baseline chest CT and extracted the whole-lung radiomics features simultaneously. Besides, baseline whole-lung CT image features are further assisted and extracted using the convolutional neural network. We used the back-propagation neural network to construct five prediction models based on different collocations of the features used for training. The area under the receiver operator characteristic curve (AUC) was used to compare the prediction performance among the five models. The Delong test was used to compare the differences in AUC between models pairwise. Results: The model integrated clinical-morphological features, whole-lung radiomic features, and whole-lung image features (CMRI) performed best among the five models, and achieved the highest AUC in the internal validation set, external test set 1, and external test set 2, which were 0.886 (95% CI: 0.841-0.921), 0.830 (95%CI: 0.749-0.893) and 0.879 (95%CI: 0.712-0.968), respectively. In the above three sets, the differences in AUC between the CMRI model and other models were significant (all P < 0.05). Moreover, the accuracy of the CMRI model in the external test set 3 was 96.88%. Conclusion: The baseline whole-lung CT features were feasible to predict the benign and malignant of GGNs, which is helpful for more refined management of GGNs.

Prediction of meningioma grade by constructing a clinical radiomics model nomogram based on magnetic resonance imaging.

PURPOSE: To explore the clinical value of a clinical radiomics model nomogram based on magnetic resonance imaging (MRI) for preoperative meningioma grading. MATERIALS AND METHODS: We collected retrospectively 544 patients with pathological diagnosis of meningiomas were categorized into training (n = 380) and validation (n = 164) groups at the ratio of 7∶ 3. There were 3,376 radiomics features extracted from T2WI and T1C by shukun technology platform after manual segmentation using an independent blind method by two radiologists. The Selectpercentile and Lasso are used to filter the most strongly correlated features. Random forest (RF) radiomics model and clinical radiomics model nomogram were constructed respectively. The calibration, discrimination, and clinical validity were evaluated by using the calibration curve and decision analysis curve (DCA). RESULTS: The RF radiomics model based on T1C and T2WI was the most effective to predict meningioma grade before surgery among the six different classifiers. The predictive ability of clinical radiomics model was slightly higher than that of RF model alone. The AUC, SEN, SPE, and ACC of the training set were 0.949, 0.976, 0.785, and 0.826, and the AUC, SEN, SPE, and ACC of the validation set were 0.838, 0.829, 0.783, and 0.793, respectively. The calibration curve and Hosmer-Lemeshow test showed the predictive probability of the fusion model was similar to the actual differentiated LGM and HGM. The analysis of the decision curve showed that the clinical radiomics model could obtain the best clinical net profit. CONCLUSIONS: The clinical radiomics model nomogram based on T1C and T2WI has high accuracy and sensitivity for predicting meningioma grade.

Baseline MRI-based radiomics model assisted predicting disease progression in nasopharyngeal carcinoma patients with complete response after treatment.

BACKGROUND: Accurate pretreatment prediction for disease progression of nasopharyngeal carcinoma is key to intensify therapeutic strategies to high-risk individuals. Our aim was to evaluate the value of baseline MRI-based radiomics machine-learning models in predicting the disease progression in nasopharyngeal carcinoma patients who achieved complete response after treatment. METHODS: In this retrospective study, 171 patients with pathologically confirmed nasopharyngeal carcinoma were included. Using hold-out cross validation scheme (7:3), relevant radiomic features were selected with the least absolute shrinkage and selection operator method based on baseline T2-weighted fat suppression and contrast-enhanced T1-weighted images in the training cohort. After Pearson's correlation analysis of selected radiomic features, multivariate logistic regression analysis was applied to radiomic features and clinical characteristics selection. Logistic regression analysis and support vector machine classifier were utilized to build the predictive model respectively. The predictive accuracy of the model was evaluated by ROC analysis along with sensitivity, specificity and AUC calculated in the validation cohort. RESULTS: A prediction model using logistic regression analysis comprising 4 radiomics features (HGLZE_T2H, HGLZE_T1, LDLGLE_T1, and GLNU_T1) and 5 clinical features (histology, T stage, N stage, smoking history, and age) showed the best performance with an AUC of 0.75 in the training cohort (95% CI: 0.66-0.83) and 0.77 in the validation cohort (95% CI: 0.64-0.90). The nine independent impact factors were entered into the nomogram. The calibration curves for probability of 3-year disease progression showed good agreement. The features of this prediction model showed satisfactory clinical utility with decision curve analysis. CONCLUSIONS: A radiomics model derived from pretreatment MR showed good performance for predicting disease progression in nasopharyngeal carcinoma and may help to improve clinical decision making.

Vertebral MRI-based radiomics model to differentiate multiple myeloma from metastases: influence of features number on logistic regression model performance.

OBJECTIVES: This study aimed to use the most frequent features to establish a vertebral MRI-based radiomics model that could differentiate multiple myeloma (MM) from metastases and compare the model performance with different features number. METHODS: We retrospectively analyzed conventional MRI (T1WI and fat-suppression T2WI) of 103 MM patients and 138 patients with metastases. The feature selection process included four steps. The first three steps defined as conventional feature selection (CFS), carried out 50 times (ten times with 5-fold cross-validation), included variance threshold, SelectKBest, and least absolute shrinkage and selection operator. The most frequent fixed features were selected for modeling during the last step. The number of events per independent variable (EPV) is the number of patients in a smaller subgroup divided by the number of radiomics features considered in developing the prediction model. The EPV values considered were 5, 10, 15, and 20. Therefore, we constructed four models using the top 16, 8, 6, and 4 most frequent features, respectively. The models constructed with features selected by CFS were also compared. RESULTS: The AUCs of 20EPV-Model, 15EPV-Model, and CSF-Model (AUC = 0.71, 0.81, and 0.78) were poor than 10EPV-Model (AUC = 0.84, p < 0.001). The AUC of 10EPV-Model was comparable with 5EPV-Model (AUC = 0.85, p = 0.480). CONCLUSIONS: The radiomics model constructed with an appropriate small number of the most frequent features could well distinguish metastases from MM based on conventional vertebral MRI. Based on our results, we recommend following the 10 EPV as the rule of thumb for feature selection. KEY POINTS: • The developed radiomics model could distinguish metastases from multiple myeloma based on conventional vertebral MRI. • An accurate model based on just a handful of the most frequent features could be constructed by utilizing multiple feature reduction techniques. • An event per independent variable value of 10 is recommended as a rule of thumb for modeling feature selection.

Prediction of High-Risk Cytogenetic Status in Multiple Myeloma Based on Magnetic Resonance Imaging: Utility of Radiomics and Comparison of Machine Learning Methods.

BACKGROUND: Radiomics has shown promising results in the diagnosis, efficacy, and prognostic assessments of multiple myeloma (MM). However, little evidence exists on the utility of radiomics in predicting a high-risk cytogenetic (HRC) status in MM. PURPOSE: To develop and test a magnetic resonance imaging (MRI)-based radiomics model for predicting an HRC status in MM patients. STUDY TYPE: Retrospective. POPULATION: Eighty-nine MM patients (HRC [n: 37] and non-HRC [n: 52]). FIELD STRENGTH/SEQUENCE: A 3.0 T; fast spin-echo (FSE): T1-weighted image (T1WI) and fat-suppression T2WI (FS-T2WI). ASSESSMENT: Overall, 1409 radiomics features were extracted from each volume of interest drawn by radiologists. Three sequential feature selection steps-variance threshold, SelectKBest, and least absolute shrinkage selection operator-were repeated 10 times with 5-fold cross-validation. Radiomics models were constructed with the top three frequency features of T1 WI/T2 WI/two-sequence MRI (T1 WI and FS-T2 WI). Radiomics models, clinical data (age and visually assessed MRI pattern), or radiomics combined with clinical data were used with six classifiers to distinguish between HRC and non-HRC statuses. Six classifiers used were support vector machine, random forest, logistic regression (LR), decision tree, k-nearest neighbor, and XGBoost. Model performance was evaluated with area under the curve (AUC) values. STATISTICAL TESTS: Mann-Whitney U-test, Chi-squared test, Z test, and DeLong method. RESULTS: The LR classifier performed better than the other classifiers based on different data (AUC: 0.65-0.82; P < 0.05). The two-sequence MRI models performed better than the other data models using different classifiers (AUC: 0.68-0.82; P < 0.05). Thus, the LR two-sequence model yielded the best performance (AUC: 0.82 ± 0.02; sensitivity: 84.1%; specificity: 68.1%; accuracy: 74.7%; P < 0.05). CONCLUSION: The LR-based machine learning method appears superior to other classifier methods for assessing HRC in MM. Radiomics features based on two-sequence MRI showed good performance in differentiating HRC and non-HRC statuses in MM. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Radiomics Nomograms Based on Non-enhanced MRI and Clinical Risk Factors for the Differentiation of Chondrosarcoma from Enchondroma.

BACKGROUND: Differentiating chondrosarcoma from enchondroma using conventional MRI remains challenging. An effective method for accurate preoperative diagnosis could affect the management and prognosis of patients. PURPOSE: To validate and evaluate radiomics nomograms based on non-enhanced MRI and clinical risk factors for the differentiation of chondrosarcoma from enchondroma. STUDY TYPE: Retrospective. POPULATION: A total of 103 patients with pathologically confirmed chondrosarcoma (n = 53) and enchondroma (n = 50) were randomly divided into training (n = 68) and validation (n = 35) groups. FIELD STRENGTH/SEQUENCE: Axial non-contrast-enhanced T1-weighted images (T1WI) and fat-suppressed T2-weighted images (T2WI-FS) were acquired at 3.0 T. ASSESSMENT: Clinical risk factors (sex, age, and tumor location) and diagnosis assessment based on morphologic MRI by three radiologists were recorded. Three radiomics signatures were established based on the T1WI, T2WI-FS, and T1WI + T2WI-FS sequences. Three clinical radiomics nomograms were developed based on the clinical risk factors and three radiomics signatures. STATISTICAL TESTS: The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of radiomics signatures and clinical radiomics nomograms. RESULTS: Tumor location was an important clinical risk factor (P < 0.05). The radiomics signature based on T1WI and T1WI + T2WI-FS features performed better than that based on T2WI-FS in the validation group (AUC in the validation group: 0.961, 0.938, and 0.833, respectively; P < 0.05). In the validation group, the three clinical radiomics nomograms (T1WI, T2WI-FS, and T1WI + T2WI-FS) achieved AUCs of 0.938, 0.935, and 0.954, respectively. In all patients, the clinical radiomics nomogram based on T2WI-FS (AUC = 0.967) performed better than that based on T2WI-FS (AUC = 0.901, P < 0.05). DATA CONCLUSION: The proposed clinical radiomics nomogram showed promising performance in differentiating chondrosarcoma from enchondroma. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY: Stage 2.

Prognostic and predictive value of radiomics features at MRI in nasopharyngeal carcinoma.

PURPOSE: To explore the value of MRI-based radiomics features in predicting risk in disease progression for nasopharyngeal carcinoma (NPC). METHODS: 199 patients confirmed with NPC were retrospectively included and then divided into training and validation set using a hold-out validation (159: 40). Discriminative radiomic features were selected with a Wilcoxon signed-rank test from tumors and normal masticatory muscles of 37 NPC patients. LASSO Cox regression and Pearson correlation analysis were applied to further confirm the differential expression of the radiomic features in the training set. Using the multiple Cox regression model, we built a radiomic feature-based classifier, Rad-Score. The prognostic and predictive performance of Rad-Score was validated in the validation cohort and illustrated in all included 199 patients. RESULTS: We identified 1832 differentially expressed radiomic features between tumors and normal tissue. Rad-Score was built based on one radiomic feature: CET1-w_wavelet.LLH_GLDM_Dependence-Entropy. Rad-Score showed a satisfactory performance to predict disease progression in NPC with an area under the curve (AUC) of 0.604, 0.732, 0.626 in the training, validation, and the combined cohort (all 199 patients included) respectively. Rad-Score improved risk stratification, and disease progression-free survival was significantly different between these groups in every cohort of patients (p = 0.044 or p < 0.01). Combining radiomics and clinical features, higher AUC was achieved of the prediction of 3-year disease progression-free survival (PFS) (AUC, 0.78) and 5-year disease PFS (AUC, 0.73), although there was no statistical difference. CONCLUSION: The radiomics classifier, Rad-Score, was proven useful for pretreatment prognosis prediction and showed potential in risk stratification for NPC. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12672-021-00460-3.

MRI-based delta-radiomics are predictive of pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer.

RATIONALE AND OBJECTIVES: To investigate the capability of delta-radiomics to predict pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer (LARC). MATERIALS AND METHODS: This retrospective study enrolled 165 consecutive patients with LARC (training set, n = 116; test set, n = 49) who received nCRT before surgery. All patients underwent pre- and post-nCRT MRI examination from which radiomics features were extracted. A delta-radiomics feature was defined as the percentage change in a radiomics feature from pre- to post-nCRT MRI. A data reduction and feature selection process including the least absolute shrinkage and selection operator algorithm was performed for building T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) delta-radiomics signature. Logistic regression was used to build a T2WI and DWI combined radiomics model. Receiver operating characteristic analysis was performed to assess diagnostic performance. Delong method was used to compare the performance of delta-radiomics model with that of magnetic resonance tumor regression grade (mrTRG). RESULTS: Twenty-seven of 165 patients (16.4%) achieved pCR. T2WI and DWI delta-radiomics signature, and the combined model showed good predictive performance for pCR. The combined model achieved the highest areas under the receiver operating characteristic curves of 0.91 (95% confidence interval: 0.85-0.98) and 0.91 (95% confidence interval: 0.83-0.99) in the training and test sets, respectively (significantly greater than those for mrTRG; training set, p < 0.001; test set, p = 0.04). CONCLUSION: MRI-based delta-radiomics can help predict pCR after nCRT in patients with LARC with better performance than mrTRG.

Preoperative prediction for pathological grade of hepatocellular carcinoma via machine learning-based radiomics.

OBJECTIVE: To investigate the efficacy of contrast-enhanced computed tomography (CECT)-based radiomics signatures for preoperative prediction of pathological grades of hepatocellular carcinoma (HCC) via machine learning. METHODS: In this single-center retrospective study, data collected from 297 consecutive subjects with HCC were allocated to training dataset (n = 237) and test dataset (n = 60). Manual segmentation of lesion sites was performed with ITK-SNAP, the radiomics features were extracted by the Pyradiomics, and radiomics signatures were synthesized using recursive feature elimination (RFE) method. The prediction models for pathological grading of HCC were established by using eXtreme Gradient Boosting (XGBoost). The performance of the models was evaluated using the AUC along with 95% confidence intervals (CIs) and standard deviation, sensitivity, specificity, and accuracy. RESULTS: The radiomics signatures were found highly efficient for machine learning to differentiate high-grade HCC from low-grade HCC. For the clinical factors, when they were merely applied to train a machine learning model, the model achieved an AUC of 0.6698, along with 95% CI and standard deviation of 0.5307-0.8089 and 0.0710, respectively (sensitivity, 0.6522; specificity, 0.4595; accuracy, 0.5333). Meanwhile, when the radiomics signatures were applied in association with clinical factors to train a machine learning model, the performance of the model remarkably increased with AUC of 0.8014, along with 95% CI and standard deviation of 0.6899-0.9129 and 0.0569, respectively (sensitivity, 0.6522; specificity, 0.7297; accuracy, 0.7000). CONCLUSIONS: The radiomics signatures could non-invasively explore the underlying association between CECT images and pathological grades of HCC. KEY POINTS: • The radiomics signatures may non-invasively explore the underlying association between CECT images and pathological grades of HCC via machine learning. • The radiomics signatures of CECT images may enhance the prediction performance of pathological grading of HCC, and further validation is required. • The features extracted from arterial phase CECT images may be more reliable than venous phase CECT images for predicting pathological grades of HCC.