Pathologic stratification of operable lung adenocarcinoma using radiomics features extracted from dual energy CT images

Enhanced CT-Based Intratumoral and Peritumoral Radiomics Nomograms Predict High-Grade Patterns of Invasive Lung Adenocarcinoma

RATIONALE AND OBJECTIVES

Extraction of intratumoral and peritumoral radiomics features combined with clinical factors to establish nomograms to predict high-grade patterns (micropapillary and solid) of invasive adenocarcinoma of the lung (IAC).

MATERIALS AND METHODS

A retrospective study was conducted on 463 patients with pathologically confirmed IAC. Patients were randomized in a 7:3 ratio into a training cohort (n = 324) and a testing cohort (n = 139). A total of 2154 CT-based radiomic features were extracted from each of the four regions: gross tumor volume (GTV) and gross peritumoral tumor volume (GPTV3, GPTV6, GPTV9) containing peri-tumor regions of 3 mm, 6 mm, and 9 mm. A radiomics nomogram was constructed based on the optimal radiomics model and clinically independent predictors.

RESULTS

The GPTV3 radiomics model showed better predictive performance in the testing group compared to the GTV (0.840), GPTV6 (0.843), and GPTV9 (0.734) models, with an AUC value of 0.889 in the testing group. In the clinical model, tumor density and the presence of a spiculation sign were identified as independent predictors. The nomogram, which combined these independent predictors with the GPTV3-Radscore, proved to be clinically useful.

CONCLUSION

The GPTV3 radiomics model was superior to the GTV, GPTV6, and GPTV9 radiomics models in predicting high-grade patterns (HGP) of IAC. In addition, nomograms based on GPTV3 radiomics features and clinically independent predictors can further improve the prediction efficiency.

Deep-learning-based 3D super-resolution CT radiomics model: Predict the possibility of the micropapillary/solid component of lung adenocarcinoma

Objective

Invasive lung adenocarcinoma（ILA） with micropapillary (MPP)/solid (SOL) components has a poor prognosis. Preoperative identification is essential for decision-making for subsequent treatment. This study aims to construct and evaluate a super-resolution（SR） enhanced radiomics model designed to predict the presence of MPP/SOL components preoperatively to provide more accurate and individualized treatment planning.

Methods

Between March 2018 and November 2023, patients who underwent curative intent ILA resection were included in the study. We implemented a deep transfer learning network on CT images to improve their resolution, resulting in the acquisition of preoperative super-resolution CT (SR-CT) images. Models were developed using radiomic features extracted from CT and SR-CT images. These models employed a range of classifiers, including Logistic Regression (LR), Support Vector Machines (SVM), k-Nearest Neighbors (KNN), Random Forest, Extra Trees, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Multilayer Perceptron (MLP). The diagnostic performance of the models was assessed by measuring the area under the curve (AUC).

Result

A total of 245 patients were recruited, of which 109 (44.5 %) were diagnosed with ILA with MPP/SOL components. In the analysis of CT images, the SVM model exhibited outstanding effectiveness, recording AUC scores of 0.864 in the training group and 0.761 in the testing group. When this SVM approach was used to develop a radiomics model with SR-CT images, it recorded AUCs of 0.904 in the training and 0.819 in the test cohorts. The calibration curves indicated a high goodness of fit, while decision curve analysis (DCA) highlighted the model's clinical utility.

Conclusion

The study successfully constructed and evaluated a deep learning（DL）-enhanced SR-CT radiomics model. This model outperformed conventional CT radiomics models in predicting MPP/SOL patterns in ILA. Continued research and broader validation are necessary to fully harness and refine the clinical potential of radiomics when combined with SR reconstruction technology.

A combination of radiomic features, clinic characteristics, and serum tumor biomarkers to predict the possibility of the micropapillary/solid component of lung adenocarcinoma

BACKGROUND

Invasive lung adenocarcinoma with MPP/SOL components has a poor prognosis and often shows a tendency to recurrence and metastasis. This poor prognosis may require adjustment of treatment strategies. Preoperative identification is essential for decision-making for subsequent treatment.

OBJECTIVE

This study aimed to preoperatively predict the probability of MPP/SOL components in lung adenocarcinomas by a comprehensive model that includes radiomics features, clinical characteristics, and serum tumor biomarkers.

DESIGN

A retrospective case control, diagnostic accuracy study.

METHODS

This study retrospectively recruited 273 patients (males: females, 130: 143; mean age ± standard deviation, 63.29 ± 10.03 years; range 21-83 years) who underwent resection of invasive lung adenocarcinoma. Sixty-one patients (22.3%) were diagnosed with lung adenocarcinoma with MPP/SOL components. Radiomic features were extracted from CT before surgery. Clinical, radiomic, and combined models were developed using the logistic regression algorithm. The clinical and radiomic signatures were integrated into a nomogram. The diagnostic performance of the models was evaluated using the area under the curve (AUC). Studies were scored according to the Radiomics Quality Score and Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis guidelines.

RESULTS

The radiomics model achieved the best AUC values of 0.858 and 0.822 in the training and test cohort, respectively. Tumor size (T_size), solid tumor size (ST_size), consolidation-to-tumor ratio (CTR), years of smoking, CYFRA 21-1, and squamous cell carcinoma antigen were used to construct the clinical model. The clinical model achieved AUC values of 0.741 and 0.705 in the training and test cohort, respectively. The nomogram showed higher AUCs of 0.894 and 0.843 in the training and test cohort, respectively.

CONCLUSION

This study has developed and validated a combined nomogram, a visual tool that integrates CT radiomics features with clinical indicators and serum tumor biomarkers. This innovative model facilitates the differentiation of micropapillary or solid components within lung adenocarcinoma and achieves a higher AUC, indicating superior predictive accuracy.

A CT-Based Radiomics Nomogram Combined with Clinic-Radiological Characteristics for Preoperative Prediction of the Novel IASLC Grading of Invasive Pulmonary Adenocarcinoma

RATIONALE AND OBJECTIVES

The novel International Association for the Study of Lung Cancer (IASLC) grading system of invasive lung adenocarcinoma (ADC) demonstrated a remarkable prognostic effect and enabled numerous patients to benefit from adjuvant chemotherapy. We sought to build a CT-based nomogram for preoperative prediction of the IASLC grading.

MATERIALS AND METHODS

This work retrospectively analyzed the CT images and clinical data of 303 patients with pathologically confirmed invasive ADC. The histological subtypes and radiological characteristics of the patients were re-evaluated. Radiomics features were extracted, and the optimal subset of features was established by ANOVA, spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO). Univariate and multivariate analyses identified the independent clinical and radiological variables. Finally, multivariate logistic regression analysis incorporated clinical, radiological, and optimal radiomics features into the nomogram. Receiver operating characteristic (ROC) curve, and accuracy were applied to assess the model's performance. Decision curve analysis (DCA), and calibration curve were applied to assess the clinical usefulness.

RESULTS

Nine selected CT image features were used to develop the radiomics model. The accuracy, precision, sensitivity, and specificity of the radiomics model outperformed the clinic-radiological model in the training and testing sets. Integrating Radscore with independent radiological characteristics showed higher prediction performance than clinic-radiological characteristics alone in the training (AUC, 0.915 vs. 0.882; DeLong, p < 0.05) and testing (AUC, 0.838 vs. 0.782; DeLong, p < 0.05) sets. Good calibration and decision curve analysis demonstrated the clinical usefulness of the nomogram.

CONCLUSION

Radiomics features effectively predict high-grade ADC. The combined nomogram may facilitate selecting patients who benefit from adjuvant treatment.

Lung dual energy CT: Impact of different technological solutions on quantitative analysis

PURPOSE

To evaluated the accuracy of spectral parameters quantification of four different CT scanners in dual energy examinations of the lung using a dedicated phantom.

METHOD

Measurements were made with different technologies of the same vendor: one dual source CT scanner (DSCT), one TwinBeam (i.e. split filter) and two sequential acquisition single source scanners (SSCT). Angular separation of Calcium and Iodine signals were calculated from scatter plots of low-kVp versus high-kVp HUs. Electron density (ρe), effective atomic number (Zeff) and Iodine concentration (Iconc) were measured using Syngo.via software. Accuracy (A) of ρe, Zeff and Iconc was evaluated as the absolute percentage difference (D%) between reference values and measured ones, while precision (P) was evaluated as the variability σ obtained by repeating the measurement with different acquisition/reconstruction settings.

RESULTS

Angular separation was significantly larger for DSCT (α = 9.7°) and for sequential SSCT (α = 9.9°) systems. TwinBeam was less performing in material separation (α = 5.0°). The lowest average A was observed for TwinBeam (A_ρe = [4.7 ± 1.0], A_Z = [9.1 ± 3.1], A_Iconc = [19.4 ± 4.4]), while the best average A was obtained for Flash (A_ρe = [1.8 ± 0.4], A_Z = [3.5 ± 0.7], A_Iconc = [7.3 ± 1.8]). TwinBeam presented inferior average P (P_ρe = [0.6 ± 0.1], P_Z = [1.1 ± 0.2], P_Iconc = [10.9 ± 4.9]), while other technologies demonstrate a comparable average.

CONCLUSIONS

Different technologies performed material separation and spectral parameter quantification with different degrees of accuracy and precision. DSCT performed better while TwinBeam demonstrated not excellent performance. Iodine concentration measurements exhibited high variability due to low Iodine absolute content in lung nodules, thus limiting its clinical usefulness in pulmonary applications.

A Combination of Radiomic Features, Imaging Characteristics, and Serum Tumor Biomarkers to Predict the Possibility of the High-Grade Subtypes of Lung Adenocarcinoma

RATIONALE AND OBJECTIVES

Lung adenocarcinomas (LADC) containing high-grade subtypes have a poorer prognosis. And some studies have shown that high-grade subtypes have been identified as an independent predictor of local recurrence in patients treated with limited resection. The aim of this study was to construct a combined model based on radiomic features, imaging characteristics and serum tumor biomarkers to predict the possibility of preoperative high-grade subtypes.

MATERIALS AND METHODS

156 patients with LADC were retrospectively recruited in this study. These patients were randomly divided into training and validation cohorts. Radiomics features and imaging characteristics were extracted from plain CT images. A nomogram was developed in a training cohort by univariate and multivariate logistic analysis, and its performance was evaluated by receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) in the training and validation cohorts.

RESULTS

A total of 1316 radiomic features were extracted from the lesions in plain chest CT images. After applying the mRMR algorithm and the LASSO regression, 4 features were retained. Based on these radiomic features, Radiomic score (Radscore) was calculated for each patient. Spiculation, air bronchogram sign, CYFRA 21-1 and Radscore had been used in the construction of the combined model. The AUC of the combined model was respectively 0.88 (95% CI, 0.82-0.95) and 0.94 (95% CI, 0.86-1.00) in the training and validation cohorts.

CONCLUSION

The combined model based on CT images and serum tumor biomarkers, can predict the high-grade subtypes of LADC in a non-invasive manner, which may influence individual treatment planning, such as the choice of surgical approach and postoperative adjuvant therapy.

Delta radiomics model for the prediction of progression-free survival time in advanced non-small-cell lung cancer patients after immunotherapy

Objective

To assess the validity of pre- and posttreatment computed tomography (CT)-based radiomics signatures and delta radiomics signatures for predicting progression-free survival (PFS) in stage III-IV non-small-cell lung cancer (NSCLC) patients after immune checkpoint inhibitor (ICI) therapy.

Methods

Quantitative image features of the largest primary lung tumours were extracted on CT-enhanced imaging at baseline (time point 0, TP0) and after the 2^nd-3^rd immunotherapy cycles (time point 1, TP1). The critical features were selected to construct TP0, TP1 and delta radiomics signatures for the risk stratification of patient survival after ICI treatment. In addition, a prediction model integrating the clinicopathologic risk characteristics and phenotypic signature was developed for the prediction of PFS.

Results

The C-index of TP0, TP1 and delta radiomics models in the training and validation cohort were 0.64, 0.75, 0.80, and 0.61, 0.68, 0.78, respectively. The delta radiomics score exhibited good accuracy for distinguishing patients with slow and rapid progression to ICI treatment. The predictive accuracy of the combined prediction model was higher than that of the clinical prediction model in both training and validation sets (P<0.05), with a C-index of 0.83 and 0.70, respectively. Additionally, the delta radiomics model (C-index of 0.86) had a higher predictive accuracy compared to PD-L1 expression (C-index of 0.50) (P<0.0001).

Conclusions

The combined prediction model including clinicopathologic characteristics (tumour anatomical classification and brain metastasis) and the delta radiomics signature could achieve the individualized prediction of PFS in ICIs-treated NSCLC patients.

Multi-lesion radiomics of PET/CT for non-invasive survival stratification and histologic tumor risk profiling in patients with lung adenocarcinoma

OBJECTIVES

This study investigates the ability of machine learning (ML) models trained on clinical data and 2-deoxy-2-[18F]fluoro-D-glucose(FDG) positron emission tomography/computed tomography (PET/CT) radiomics to predict overall survival (OS), tumor grade (TG), and histologic growth pattern risk (GPR) in lung adenocarcinoma (LUAD) patients.

METHODS

A total of 421 treatment-naive patients with histologically-proven LUAD and available FDG PET/CT imaging were retrospectively included. Four cohorts were assessed for predicting 4-year OS (n = 276), 3-year OS (n = 280), TG (n = 298), and GPR (n = 265). FDG-avid lesions were delineated, and 2082 radiomics features were extracted and combined with endpoint-specific clinical parameters. ML models were built for the prediction of 4-year OS (M4OS), 3-year OS (M3OS), tumor grading (MTG), and histologic growth pattern risk (MGPR). A 100-fold Monte Carlo cross-validation with 80:20 training to validation split was employed as a performance evaluation for all models. The association between the M4OS and M3OS predictions with OS was assessed by the Kaplan-Meier survival analysis.

RESULTS

The area under the receiver operator characteristics curve (AUC) was the highest for M4OS (AUC 0.88, 95% confidence interval (CI) 86.7-88.7), followed by M3OS (AUC 0.84, CI 82.9-84.9), while MTG and MGPR performed equally well (AUC 0.76, CI 74.4-77.9, CI 74.6-78, respectively). Predictions of M4OS (hazard ratio (HR) -2.4, CI -2.47 to -1.64, p < 0.05) and M3OS (HR -2.36, CI -2.79 to -1.93, p < 0.05) were independently associated with OS.

CONCLUSION

ML models are able to predict long-term survival outcomes in LUAD patients with high accuracy. Furthermore, histologic grade and predominant growth pattern risk can be predicted with satisfactory accuracy.

KEY POINTS

• Machine learning models trained on pre-therapeutic PET/CT radiomics enable highly accurate long-term survival prediction of patients with lung adenocarcinoma. • Highly accurate survival predictions are achieved in lung adenocarcinoma patients despite heterogenous histologies and treatment regimens. • Radiomic machine learning models are able to predict lung adenocarcinoma tumor grade and histologic growth pattern risk with satisfactory accuracy.

Radiomic Analysis of Pulmonary Nodules for Distinguishing Malignancy From Benignancy: The Value of Using Iodine Maps From Dual-Energy Computed Tomography

OBJECTIVE

The aim of the study is to investigate the diagnostic accuracy of radiomics on iodine maps from dual-energy computed tomography (DECT) in distinguishing lung cancer from benign pulmonary nodules.

METHODS

This retrospective study was approved by the institutional review board, and written informed consent was waived. A total of 109 patients with 55 malignant nodules and 62 benign nodules underwent contrast-enhanced DECT. Eight iodine uptake parameters on iodine maps generated by DECT were calculated and established a predictive model. Eighty-seven radiomics features of entire tumor were extracted from iodine maps and established a radiomics model. The iodine uptake model and radiomics model were independently built based on the highly reproducible features using the least absolute shrinkage and selection operator method. The diagnostic accuracy of 2 models were assessed using receiver operating curve analysis. For external validation, 47 patients (25 benign and 22 malignant) from another hospital were assigned to testing data set.

RESULTS

All iodine uptake features showed significant association with malignancy (P < 0.01) and 2 selected features (mean value of virtual noncontrast images and mean value of vital part on contrast-enhanced image) constituted the iodine model. The radiomics model comprised 2 features (original shape sphericity and original glszm small area high gray level emphasis), which showed good discrimination both in the training cohort (area under the curve, 0.957) and validation cohort (area under the curve, 0.800). Radiomics model showed superior performance than iodine uptake model (accuracy, 89.7% vs 80.6%).

CONCLUSIONS

Radiomics model extracted from iodine maps provided a robust diagnostic tool for discriminating pulmonary malignant nodules and had high potential in clinical application.

Combined whole-lesion radiomic and iodine analysis for differentiation of pulmonary tumors

Quantitative radiomic and iodine imaging features have been explored for diagnosis and characterization of tumors. In this work, we invistigate combined whole-lesion radiomic and iodine analysis for the differentiation of pulmonary tumors on contrast-enhanced dual-energy CT (DECT) chest images. 100 biopsy-proven solid lung lesions on contrast-enhanced DECT chest exams within 3 months of histopathologic sampling were identified. Lesions were volumetrically segmented using open-source software. Lesion segmentations and iodine density volumes were loaded into a radiomics prototype for quantitative analysis. Univariate analysis was performed to determine differences in volumetric iodine concentration (mean, median, maximum, minimum, 10th percentile, 90th percentile) and first and higher order radiomic features (n = 1212) between pulmonary tumors. Analyses were performed using a 2-sample t test, and filtered for false discoveries using Benjamini–Hochberg method. 100 individuals (mean age 65 ± 13 years; 59 women) with 64 primary and 36 metastatic lung lesions were included. Only one iodine concentration parameter, absolute minimum iodine, significantly differed between primary and metastatic pulmonary tumors (FDR-adjusted p = 0.015, AUC 0.69). 310 (FDR-adjusted p = 0.0008 to p = 0.0491) radiomic features differed between primary and metastatic lung tumors. Of these, 21 features achieved AUC ≥ 0.75. In subset analyses of lesions imaged by non-CTPA protocol (n = 72), 191 features significantly differed between primary and metastatic tumors, 19 of which achieved AUC ≥ 0.75. In subset analysis of tumors without history of prior treatment (n = 59), 40 features significantly differed between primary and metastatic tumors, 11 of which achieved AUC ≥ 0.75. Volumetric radiomic analysis provides differentiating capability beyond iodine quantification. While a high number of radiomic features differentiated primary versus metastatic pulmonary tumors, fewer features demonstrated good individual discriminatory utility.

Can radiomic analysis of a single-phase dual-energy CT improve the diagnostic accuracy of differentiating enhancing from non-enhancing small renal lesions?

BACKGROUND

The value of dual-energy computed tomography (DECT)-based radiomics in renal lesions is unknown.

PURPOSE

To develop DECT-based radiomic models and assess their incremental values in comparison to conventional measurements for differentiating enhancing from non-enhancing small renal lesions.

MATERIAL AND METHODS

A total of 349 patients with 519 small renal lesions (390 non-enhancing, 129 enhancing) who underwent contrast-enhanced nephrographic phase DECT examinations between June 2013 and January 2020 on multiple DECT platforms were retrospectively recruited. Cohort A included all lesions, while cohort B included Bosniak II-IV and solid enhancing renal lesions. Radiomic models were built with features selected by the least absolute shrinkage and selection operator regression (LASSO). ROC analyses were performed to compare the diagnostic accuracy among conventional and radiomic models for predicting enhancing renal lesions.

RESULTS

The individual iodine concentration (IC), normalized IC, mean attenuation on 75-keV images, radiomic model of iodine images, 75-keV images and a combined model integrating all the above-mentioned features all demonstrated high AUCs for predicting renal lesion enhancement in cohort A (AUCs = 0.934-0.979) as well as in the test dataset (AUCs = 0.892-0.962) of cohort B (P values with Bonferroni correction >0.003). The AUC (0.864) of mean attenuation on 75-keV images was significantly lower than those of other models (all P values ≤0.001) except the radiomic model of 75-keV images (P = 0.038) in the training dataset of cohort B.

CONCLUSION

No incremental value was found by adding radiomic and machine learning analyses to iodine images for differentiating enhancing from non-enhancing renal lesions.

Parameter tuning in machine learning based on radiomics biomarkers of lung cancer

BACKGROUND

Lung cancer is one of the most common cancers, and early diagnosis and intervention can improve cancer cure rate.

OBJECTIVE

To improve predictive performance of radiomics features for lung cancer by tuning the machine learning model parameters.

METHODS

Using a dataset involving 263 cases (125 benign and 138 malignant) acquired from our hospital, each classifier model is trained and tested using 237 and 26 cases, respectively. We initially extract 867 radiomics features of CT images for model development and then test 10 feature selections and 7 models to determine the best method. We further tune the parameter of the final model to reach the best performance. The adjusted final model is then validated using 224 cases acquired from Lung Image Database Consortium (LIDC) dataset (64 benign and 160 malignant) with the same set of selected radiomics features.

RESULTS

During model development, the feature selection via concave minimization method show the best performance of area under ROC curve (AUC = 0.765), followed by l0-norm regularization (AUC = 0.741) and Fisher discrimination criterion (AUC = 0.734). Support vector machine (SVM) and random forest (RF) are the top two machine learning algorithms showing the best performance (AUC = 0.765 and 0.734, respectively), using by the default parameter. After parameter tuning, SVM with linear kernel achieves the best performance (AUC = 0.837), whereas the best tuned RF with the number of trees is 510 and yields a slightly lower performance (AUC = 0.775) in 26 test samples data. During model validation, the SVM and RF models yield AUC = 0.78 and 0.77, respectively.

CONCLUSION

Appropriate quantitative radiomics features and accurate parameters can improve the model's performance to predict lung cancer.

CT Texture Analysis of Pulmonary Neuroendocrine Tumors-Associations with Tumor Grading and Proliferation

Texture analysis derived from computed tomography (CT) might be able to provide clinically relevant imaging biomarkers and might be associated with histopathological features in tumors. The present study sought to elucidate the possible associations between texture features derived from CT images with proliferation index Ki-67 and grading in pulmonary neuroendocrine tumors. Overall, 38 patients (n = 22 females, 58%) with a mean age of 60.8 ± 15.2 years were included into this retrospective study. The texture analysis was performed using the free available Mazda software. All tumors were histopathologically confirmed. In discrimination analysis, "S(1,1)SumEntrp" was significantly different between typical and atypical carcinoids (mean 1.74 ± 0.11 versus 1.79 ± 0.14, p = 0.007). The correlation analysis revealed a moderate positive association between Ki-67 index with the first order parameter kurtosis (r = 0.66, p = 0.001). Several other texture features were associated with the Ki-67 index, the highest correlation coefficient showed "S(4,4)InvDfMom" (r = 0.59, p = 0.004). Several texture features derived from CT were associated with the proliferation index Ki-67 and might therefore be a valuable novel biomarker in pulmonary neuroendocrine tumors. "Sumentrp" might be a promising parameter to aid in the discrimination between typical and atypical carcinoids.

A Machine learning model trained on dual-energy CT radiomics significantly improves immunotherapy response prediction for patients with stage IV melanoma

BACKGROUND

To assess the additive value of dual-energy CT (DECT) over single-energy CT (SECT) to radiomics-based response prediction in patients with metastatic melanoma preceding immunotherapy.

MATERIAL AND METHODS

A total of 140 consecutive patients with melanoma (58 female, 63±16 years) for whom baseline DECT tumor load assessment revealed stage IV and who were subsequently treated with immunotherapy were included. Best response was determined using the clinical reports (81 responders: 27 complete response, 45 partial response, 9 stable disease). Individual lesion response was classified manually analogous to RECIST 1.1 through 1291 follow-up examinations on a total of 776 lesions (6.7±7.2 per patient). The patients were sorted chronologically into a study and a validation cohort (each n=70). The baseline DECT was examined using specialized tumor segmentation prototype software, and radiomic features were analyzed for response predictors. Significant features were selected using univariate statistics with Bonferroni correction and multiple logistic regression. The area under the receiver operating characteristic curve of the best subset was computed (AUROC). For each combination (SECT/DECT and patient response/lesion response), an individual random forest classifier with 10-fold internal cross-validation was trained on the study cohort and tested on the validation cohort to confirm the predictive performance.

RESULTS

We performed manual RECIST 1.1 response analysis on a total of 6533 lesions. Multivariate statistics selected significant features for patient response in SECT (min. brightness, R²=0.112, padj. ≤0.001) and DECT (textural coarseness, R²=0.121, padj. ≤0.001), as well as lesion response in SECT (mean absolute voxel intensity deviation, R²=0.115, padj. ≤0.001) and DECT (iodine uptake metrics, R²≥0.12, padj. ≤0.001). Applying the machine learning models to the validation cohort confirmed the additive predictive power of DECT (patient response AUROC SECT=0.5, DECT=0.75; lesion response AUROC SECT=0.61, DECT=0.85; p<0.001).

CONCLUSION

The new method of DECT-specific radiomic analysis provides a significant additive value over SECT radiomics approaches for response prediction in patients with metastatic melanoma preceding immunotherapy, especially on a lesion-based level. As mixed tumor response is not uncommon in metastatic melanoma, this lends a powerful tool for clinical decision-making and may potentially be an essential step toward individualized medicine.

Virtual Monoenergetic Images of Dual-Energy CT-Impact on Repeatability, Reproducibility, and Classification in Radiomics

Simple Summary

Virtual monoenergetic images from dual-energy CT are incrementally used in routine clinical practice. Thus, radiomic analysis will be more often performed on these images in the future. This study characterized the test–retest repeatability and reproducibility of radiomic features from virtual monoenergetic images and their impact on machine-learning-based lesion classification. The results of this study provide a basis to improve radiomic analyses and identify the role of feature stability in classification tasks when using virtual monoenergetic imaging with different scan or reconstruction parameters in multicenter clinical studies.

Abstract

The purpose of this study was to (i) evaluate the test–retest repeatability and reproducibility of radiomic features in virtual monoenergetic images (VMI) from dual-energy CT (DECT) depending on VMI energy (40, 50, 75, 120, 190 keV), radiation dose (5 and 15 mGy), and DECT approach (dual-source and split-filter DECT) in a phantom (ex vivo), and (ii) to assess the impact of VMI energy and feature repeatability on machine-learning-based classification in vivo in 72 patients with 72 hypodense liver lesions. Feature repeatability and reproducibility were determined by concordance–correlation–coefficient (CCC) and dynamic range (DR) ≥0.9. Test–retest repeatability was high within the same VMI energies and scan conditions (percentage of repeatable features ranging from 74% for SFDE mode at 40 keV and 15 mGy to 86% for DSDE at 190 keV and 15 mGy), while reproducibility varied substantially across different VMI energies and DECTs (percentage of reproducible features ranging from 32.8% for SFDE at 5 mGy comparing 40 with 190 keV to 99.2% for DSDE at 15 mGy comparing 40 with 50 keV). No major differences were observed between the two radiation doses (<10%) in all pair-wise comparisons. In vivo, machine learning classification using penalized regression and random forests resulted in the best discrimination of hemangiomas and metastases at low-energy VMI (40 keV), and for cysts at high-energy VMI (120 keV). Feature selection based on feature repeatability did not improve classification performance. Our results demonstrate the high repeatability of radiomics features when keeping scan and reconstruction conditions constant. Reproducibility diminished when using different VMI energies or DECT approaches. The choice of optimal VMI energy improved lesion classification in vivo and should hence be adapted to the specific task.

Robustness of dual-energy CT-derived radiomic features across three different scanner types

OBJECTIVES

To investigate the robustness of radiomic features between three dual-energy CT (DECT) systems.

METHODS

An anthropomorphic body phantom was scanned on three different DECT scanners, a dual-source (dsDECT), a rapid kV-switching (rsDECT), and a dual-layer detector DECT (dlDECT). Twenty-four patients who underwent abdominal DECT examinations on each of the scanner types during clinical follow-up were retrospectively included (n = 72 examinations). Radiomic features were extracted after standardized image processing, following ROI placement in phantom tissues and healthy appearing hepatic, splenic and muscular tissue of patients using virtual monoenergetic images at 65 keV (VMI_65keV) and virtual unenhanced images (VUE). In total, 774 radiomic features were extracted including 86 original features and 8 wavelet transformations hereof. Concordance correlation coefficients (CCC) and analysis of variances (ANOVA) were calculated to determine inter-scanner robustness of radiomic features with a CCC of ≥ 0.9 deeming a feature robust.

RESULTS

None of the phantom-derived features attained the threshold for high feature robustness for any inter-scanner comparison. The proportion of robust features obtained from patients scanned on all three scanners was low both in VMI_65keV (dsDECT vs. rsDECT:16.1% (125/774), dlDECT vs. rsDECT:2.5% (19/774), dsDECT vs. dlDECT:2.6% (20/774)) and VUE (dsDECT vs. rsDECT:11.1% (86/774), dlDECT vs. rsDECT:2.8% (22/774), dsDECT vs. dlDECT:2.7% (21/774)). The proportion of features without significant differences as per ANOVA was higher both in patients (51.4-71.1%) and in the phantom (60.6-73.4%).

CONCLUSIONS

The robustness of radiomic features across different DECT scanners in patients was low and the few robust patient-derived features were not reflected in the phantom experiment. Future efforts should aim to improve the cross-platform generalizability of DECT-derived radiomics.

KEY POINTS

• Inter-scanner robustness of dual-energy CT-derived radiomic features was on a low level in patients who underwent clinical examinations on three DECT platforms. • The few robust patient-derived features were not confirmed in our phantom experiment. • Limited inter-scanner robustness of dual-energy CT derived radiomic features may impact the generalizability of models built with features from one particular dual-energy CT scanner type.

Improving the accuracy of prognosis for clinical stage I solid lung adenocarcinoma by radiomics models covering tumor per se and peritumoral changes on CT

OBJECTIVES

To assess methods to improve the accuracy of prognosis for clinical stage I solid lung adenocarcinoma using radiomics based on different volumes of interests (VOIs).

METHODS

This retrospective study included patients with postoperative clinical stage I solid lung adenocarcinoma from two hospitals, center 1 and center 2. Three databases were generated: dataset A (training set from center 1), dataset B (internal test set from center 1), and dataset C (external validation test from center 2). Disease-free survival (DFS) data were collected. CT radiomics models were constructed based on four VOIs: gross tumor volume (GTV), 3 mm external to the tumor border (peritumoral volume [PTV]_0~+3), 6 mm crossing tumor border (PTV_-3~+3), and 6 mm external to the tumor border (PTV_0~+6). The area under the receiver operating characteristic curve (AUC) was used to compare the model accuracies.

RESULTS

A total of 334 patients were included (204 and 130 from centers 1 and 2). The model using PTV_-3~+3 (AUC 0.81 [95% confidence interval {CI}: 0.75, 0.94], 0.81 [0.63, 0.90] for datasets B and C) outperformed the other three models, GTV (0.73 [0.58, 0.81], 0.73 [0.58, 0.83]), PTV_0~+3 (0.76 [0.52, 0.87], 0.75 [0.60, 0.83]), and PTV_0~+6 (0.72 [0.60, 0.81], 0.69 [0.59, 0.81]), in datasets B and C, all p < 0.05.

CONCLUSIONS

A radiomics model based on a VOI of 6 mm crossing tumor border more accurately predicts prognosis of clinical stage I solid lung adenocarcinoma than that based on VOIs including overall tumor or external rims of 3 mm and 6 mm.

KEY POINTS

• Radiomics is a useful approach to improve the accuracy of prognosis for stage I solid adenocarcinoma. • The radiomics model based on VOIs that includes 3 mm within and external to the tumor border (peritumoral volume [PTV]_-3~+3) outperformed models that included either only the tumor itself or those that only included the peritumoral volume.

CT texture analysis of tonsil cancer: Discrimination from normal palatine tonsils

The purposes of the study were to determine whether there are differences in texture analysis parameters between tonsil cancers and normal tonsils, and to correlate texture analysis with 18F-FDG PET/CT to investigate the relationship between texture analysis and metabolic parameters. Sixty-four patients with squamous cell carcinoma of the palatine tonsil were included. A ROI was drawn, including all slices, to involve the entire tumor. The contralateral normal tonsil was used for comparison with the tumors. Texture analysis parameters, mean, standard deviation (SD), entropy, mean positive pixels, skewness, and kurtosis were obtained using commercially available software. Parameters were compared between the tumor and the normal palatine tonsils. Comparisons were also performed among early tonsil cancer, advanced tonsil cancer, and normal tonsils. An ROC curve analysis was performed to assess discrimination of tumor from normal tonsils. Correlation between texture analysis and 18F-FDG PET/CT was performed. Compared to normal tonsils, the tumors showed a significantly lower mean, higher SD, higher entropy, lower skewness, and higher kurtosis on most filters (p<0.001). On comparisons among normal tonsils, early cancers, and advanced tonsil cancers, SD and entropy showed significantly higher values on all filters (p<0.001) between early cancers and normal tonsils. The AUC from the ROC analysis was 0.91, obtained from the entropy. A mild correlation was shown between texture parameters and metabolic parameters. The texture analysis parameters, especially entropy, showed significant differences in contrast-enhanced CT results between tumor and normal tonsils, and between early tonsil cancers and normal tonsils. Texture analysis can be useful as an adjunctive tool for the diagnosis of tonsil cancers.

CT-Based Radiomics Signature: A Potential Biomarker for Predicting Postoperative Recurrence Risk in Stage II Colorectal Cancer

Objective: To evaluate whether a radiomics signature could improve stratification of postoperative risk and prediction of chemotherapy benefit in stage II colorectal cancer (CRC) patients.

Material and Methods: This retrospective study enrolled 299 stage II CRC patients from January 2010 to December 2015. Based on preoperative portal venous-phase CT scans, radiomics features were generated and selected to build a radiomics score (Rad-score) using the Least Absolute Shrinkage and Selection Operator (LASSO) method. The minority group was balanced by the synthetic minority over-sampling technique (SMOTE). Predictive models were built with the Rad-score and clinicopathological factors, and the area under the curve (AUC) was used to evaluate their performance. A nomogram was also constructed for predicting 3-year disease-free survival (DFS). The performance of the nomogram was assessed with a concordance index (C-index) and calibration plots.

Results: Overall, 114 features were selected to construct the Rad-score, which was significantly associated with the 3-year DFS. Multivariate analysis demonstrated that the Rad-score, CA724 level, mismatch repair status, and perineural invasion were independent predictors of recurrence. Results showed that the Rad-score can classify patients into high-risk and low-risk groups in the training cohort (AUC 0.886) and the validation cohort (AUC 0.874). On this basis, a nomogram that integrated the Rad-score and clinical variables demonstrated superior performance (AUC 0.954, 0.906) than the clinical model alone (AUC 0.765, 0.705) in the training and validation cohorts, respectively. The C-index of the nomogram was 0.872, and the performance was acceptable.

Conclusion: Our radiomics-based model can reliably predict recurrence risk in stage II CRC patients and potentially provide complementary prognostic value to the traditional clinicopathological risk factors for better identification of patients who are most likely to benefit from adjuvant therapy. The proposed nomogram promises to be an effective tool for personalized postoperative surveillance for stage II CRC patients.

Quantitative imaging decision support (QIDSTM) tool consistency evaluation and radiomic analysis by means of 594 metrics in lung carcinoma on chest CT scan

Objective:

To evaluate the consistency of the quantitative imaging decision support (QIDS^TM) tool and radiomic analysis using 594 metrics in lung carcinoma on chest CT scan.

Materials and Methods:

We included, retrospectively, 150 patients with histologically confirmed lung cancer who underwent chemotherapy and baseline and follow-ups CT scans. Using the QIDS^TM platform, 3 radiologists segmented each lesion and automatically collected the longest diameter and the density mean value. Inter-observer variability, Bland Altman analysis and Spearman’s correlation coefficient were performed. QIDS^TM tool consistency was assessed in terms of agreement rate in the treatment response classification. Kruskal Wallis test and the least absolute shrinkage and selection operator (LASSO) method with 10-fold cross validation were used to identify radiomic metrics correlated with lesion size change.

Results:

Good and significant correlation was obtained between the measurements of largest diameter and of density among the QIDS^TM tool and the radiologists measurements. Inter-observer variability values were over 0.85. HealthMyne QIDS^TM tool quantitative volumetric delineation was consistent and matched with each radiologist measurement considering the RECIST classification (80-84%) while a lower concordance among QIDS^TM and the radiologists CHOI classification was observed (58-63%). Among 594 extracted metrics, significant and robust predictors of RECIST response were energy, histogram entropy and uniformity, Kurtosis, coronal long axis, longest planar diameter, surface, Neighborhood Grey-Level Different Matrix (NGLDM) dependence nonuniformity and low dependence emphasis as Volume, entropy of Log(2.5 mm), wavelet energy, deviation and root man squared.

Conclusion:

In conclusion, we demonstrated that HealthMyne quantitative volumetric delineation was consistent and that several radiomic metrics extracted by QIDS^TM were significant and robust predictors of RECIST response.

Prediction of micropapillary and solid pattern in lung adenocarcinoma using radiomic values extracted from near-pure histopathological subtypes

OBJECTIVES

Near-pure lung adenocarcinoma (ADC) subtypes demonstrate strong stratification of radiomic values, providing basic information for pathological subtyping. We sought to predict the presence of high-grade (micropapillary and solid) components in lung ADCs using quantitative image analysis with near-pure radiomic values.

METHODS

Overall, 103 patients with lung ADCs of various histological subtypes were enrolled for 10-repetition, 3-fold cross-validation (cohort 1); 55 were enrolled for testing (cohort 2). Histogram and textural features on computed tomography (CT) images were assessed based on the "near-pure" pathological subtype data. Patch-wise high-grade likelihood prediction was performed for each voxel within the tumour region. The presence of high-grade components was then determined based on a volume percentage threshold of the high-grade likelihood area. To compare with quantitative approaches, consolidation/tumour (C/T) ratio was evaluated on CT images; we applied radiological invasiveness (C/T ratio > 0.5) for the prediction.

RESULTS

In cohort 1, patch-wise prediction, combined model (C/T ratio and patch-wise prediction), whole-lesion-based prediction (using only the "near-pure"-based prediction model), and radiological invasiveness achieved a sensitivity and specificity of 88.00 ± 2.33% and 75.75 ± 2.82%, 90.00 ± 0.00%, and 77.12 ± 2.67%, 66.67% and 90.41%, and 90.00% and 45.21%, respectively. The sensitivity and specificity, respectively, for cohort 2 were 100.0% and 95.35% using patch-wise prediction, 100.0% and 95.35% using combined model, 75.00% and 95.35% using whole-lesion-based prediction, and 100.0% and 69.77% using radiological invasiveness.

CONCLUSION

Using near-pure radiomic features and patch-wise image analysis demonstrated high levels of sensitivity and moderate levels of specificity for high-grade ADC subtype-detecting.

KEY POINTS

• The radiomic values extracted from lung adenocarcinoma with "near-pure" histological subtypes provide useful information for high-grade (micropapillary and solid) components detection. • Using near-pure radiomic features and patch-wise image analysis, high-grade components of lung adenocarcinoma can be predicted with high sensitivity and moderate specificity. • Using near-pure radiomic features and patch-wise image analysis has potential role in facilitating the prediction of the presence of high-grade components in lung adenocarcinoma prior to surgical resection.

Radiomics predict postoperative survival of patients with primary liver cancer with different pathological types

Background

Radiomics can be used to determine the prognosis of liver cancer, but it might vary among cancer types. This study aimed to explore the clinicopathological features, radiomics, and survival of patients with hepatocellular carcinoma (HCC), mass-type cholangiocarcinoma (MCC), and combined hepatocellular-cholangiocarcinoma (CHCC).

Methods

This was a retrospective cohort study of patients with primary liver cancer operated at the department of hepatobiliary surgery of the First Affiliated Hospital of Zhejiang University from 07/2013 to 11/2015. All patients underwent preoperative liver enhanced MRI scans and diffusion-weighted imaging (DWI). The radiomics characteristics of DWI and the enhanced equilibrium phase (EP) images were extracted. The mRMR (minimum redundancy maximum relevance) was applied to filter the parameters.

Results

There were 44 patients with MCC, 59 with HCC, and 33 with CHCC. Macrovascular invasion, tumor diameter, positive ferritin preoperatively, positive AFP preoperatively, positive CEA preoperatively, Correlation, Inverse Difference Moment, and Cluster Prominence in model A (DWI and clinicopathological parameters) were independently associated with overall survival (OS) (P<0.05). Lymphadenopathy, gender, positive ferritin preoperatively, positive AFP preoperatively, positive CEA preoperatively, Uniformity, and Cluster Prominence in model B (EP and clinicopathological parameters) were independently associated with OS (P<0.05). Macrovascular invasion, lymphadenopathy, gender, positive ferritin preoperatively, positive CEA preoperatively, Uniformity_EP, GLCMEnergy_DWI, and Cluster Prominence_EP in model C (image texture and clinicopathological parameters) were independently associated with OS (P<0.05). Those factors were used to construct three nomograms to predict OS.

Conclusions

Clinicopathological and radiomics features are independently associated with the OS of patients with primary liver cancer.

The development and validation of a radiomic nomogram for the preoperative prediction of lung adenocarcinoma

Background

Accurate diagnosis of early lung cancer from small pulmonary nodules (SPN) is challenging in clinical setting. We aimed to develop a radiomic nomogram to differentiate lung adenocarcinoma from benign SPN.

Methods

This retrospective study included a total of 210 pathologically confirmed SPN (≤ 10 mm) from 197 patients, which were randomly divided into a training dataset (n = 147; malignant nodules, n = 94) and a validation dataset (n = 63; malignant nodules, n = 39). Radiomic features were extracted from the cancerous volumes of interest on contrast-enhanced CT images. The least absolute shrinkage and selection operator (LASSO) regression was used for data dimension reduction, feature selection, and radiomic signature building. Using multivariable logistic regression analysis, a radiomic nomogram was developed incorporating the radiomic signature and the conventional CT signs observed by radiologists. Discrimination and calibration of the radiomic nomogram were evaluated.

Results

The radiomic signature consisting of five radiomic features achieved an AUC of 0.853 (95% confidence interval [CI]: 0.735–0.970), accuracy of 81.0%, sensitivity of 82.9%, and specificity of 77.3%. The two conventional CT signs achieved an AUC of 0.833 (95% CI: 0.707–0.958), accuracy of 65.1%, sensitivity of 53.7%, and specificity of 86.4%. The radiomic nomogram incorporating the radiomic signature and conventional CT signs showed an improved AUC of 0.857 (95% CI: 0.723–0.991), accuracy of 84.1%, sensitivity of 85.4%, and specificity of 81.8%. The radiomic nomogram had good calibration power.

Conclusion

The radiomic nomogram might has the potential to be used as a non-invasive tool for individual prediction of SPN preoperatively. It might facilitate decision-making and improve the management of SPN in the clinical setting.

Semiautomatic Segmentation and Radiomics for Dual-Energy CT: A Pilot Study to Differentiate Benign and Malignant Hepatic Lesions

OBJECTIVE. This study assessed a machine learning-based dual-energy CT (DECT) tumor analysis prototype for semiautomatic segmentation and radiomic analysis of benign and malignant liver lesions seen on contrast-enhanced DECT. MATERIALS AND METHODS. This institutional review board-approved study included 103 adult patients (mean age, 65 ± 15 [SD] years; 53 men, 50 women) with benign (60/103) or malignant (43/103) hepatic lesions on contrast-enhanced dual-source DECT. Most malignant lesions were histologically proven; benign lesions were either stable on follow-up CT or had characteristic benign features on MRI. Low- and high-kilovoltage datasets were deidentified, exported offline, and processed with the DECT tumor analysis for semiautomatic segmentation of the volume and rim of each liver lesion. For each segmentation, contrast enhancement and iodine concentrations as well as radiomic features were derived for different DECT image series. Statistical analyses were performed to determine if DECT tumor analysis and radiomics can differentiate benign from malignant liver lesions. RESULTS. Normalized iodine concentration and mean iodine concentration in the benign and malignant lesions were significantly different (p < 0.0001-0.0084; AUC, 0.695-0.856). Iodine quantification and radiomic features from lesion rims (AUC, ≤ 0.877) had higher accuracy for differentiating liver lesions compared with the values from lesion volumes (AUC, ≤ 0.856). There was no difference in the accuracies of DECT iodine quantification (AUC, 0.91) and radiomics (AUC, 0.90) for characterizing liver lesions. CONCLUSION. DECT radiomics were more accurate than iodine quantification for differentiating solid benign and malignant hepatic lesions.

A Single Enhanced Dual-Energy CT Scan May Distinguish Lung Squamous Cell Carcinoma From Adenocarcinoma During the Venous phase

RATIONALE AND OBJECTIVES

To investigate whether iodine quantification extracted from enhanced dual energy-computed tomography (DE-CT) is useful for distinguishing lung squamous cell carcinoma from adenocarcinoma and to evaluate whether a single scan evaluated during the venous phase (VP) can be substituted for scans evaluated during other phases.

MATERIALS AND METHODS

Sixty-two patients with lung cancer (32 squamous cell carcinomas; 30 adenocarcinomas) underwent enhanced dual-phase DE-CT scans, including an arterial phase and VP. The iodine concentration (IC), normalized iodine concentration (NIC), and slope of the curve (K) in lesions were measured during two scanning phases in two different pathological types of lung cancers. The differences in parameters (IC, NIC, and K) between these two types of lung cancers were statistically analyzed. In addition, the receiver operating characteristic curves of these parameters were performed to discriminate squamous cell carcinoma from adenocarcinoma.

RESULTS

The mean IC, NIC, and K in adenocarcinomas were all higher than those in squamous cell carcinomas during the two scanning phases. However, the differences in these parameters between the two types of cancers were significant only during the VP, not during the arterial phase. Receiver operating characteristic analysis demonstrated that the optimal thresholds of the IC, NIC, and K for discriminating squamous cell carcinoma from adenocarcinoma were 1.550, 0.227, and 1.608, respectively. In addition, the sensitivity, specificity, and area under the curve were 81.2%, 83.3%, and 0.871 for the IC; 56.2%, 93.3%, and 0.800 for the NIC; and 65.6%, 80%, and 0.720 for the K; 81.3%, 83.3%, and 0.874 for the IC + NIC; 68.8%, 93.3%, and 0.891 for the IC + NIC + K, respectively. The "IC + NIC + K" had the highest diagnostic efficiency for discriminating two types of lung cancers, but with low sensitivity. Whereas, "IC"and "IC + NIC" had the similar lower diagnostic efficiency, but with high sensitivity and specificity.

CONCLUSION

The iodine quantification parameters derived from enhanced DE-CT during the VP may be useful for distinguishing lung squamous cell carcinoma from adenocarcinoma.

Differentiation of predominant subtypes of lung adenocarcinoma using a quantitative radiomics approach on CT

OBJECTIVES

To develop a model for differentiating the predominant subtype-based prognostic groups of lung adenocarcinoma using CT radiomic features, and to validate its performance in comparison with radiologists' assessments.

METHODS

A total of 993 patients presenting with invasive lung adenocarcinoma between March 2010 and June 2016 were identified. Predominant histologic subtypes were categorized into three groups according to their prognosis (group 0: lepidic; group 1: acinar/papillary; group 2: solid/micropapillary). Seven hundred eighteen radiomic features were extracted from segmented lung cancers on contrast-enhanced CT. A model-development set was formed from the images of 893 patients, while 100 image sets were reserved for testing. A least absolute shrinkage and selection operator method was used for feature selection. Performance of the radiomic model was evaluated using receiver operating characteristic curve analysis, and accuracy on the test set was compared with that of three radiologists with varying experiences (6, 7, and 19 years in chest CT).

RESULTS

Our model differentiated the three groups with areas under the curve (AUCs) of 0.892 and 0.895 on the development and test sets, respectively. In pairwise discrimination, the AUC was highest for group 0 vs. 2 (0.984). The accuracy of the model on the test set was higher than the averaged accuracy of the three radiologists without statistical significance (73.0% vs. 61.7%, p = 0.059). For group 2, the model achieved higher PPV than the observers (85.7% vs. 35.0-48.4%).

CONCLUSIONS

Predominant subtype-based prognostic groups of lung adenocarcinoma were classified by a CT-based radiomic model with comparable performance to radiologists.

KEY POINTS

• A CT-based radiomic model differentiated three prognosis-based subtype groups of lung adenocarcinoma with areas under the curve (AUCs) of 0.892 and 0.895 on development and test sets, respectively. • The CT-based radiomic model showed near perfect discrimination between group 0 and group 2 (AUCs, 0.984-1.000). • The accuracy of the CT-based radiomic model was comparable to the averaged accuracy of the three radiologists with 6, 7, and 19 years of clinical experience in chest CT (73.0% vs. 61.7%, p = 0.059), achieving a higher positive predictive value for group 2 than the observers (85.7% vs. 35.0-48.4%).

Texture Analysis: An Emerging Clinical Tool for Pancreatic Lesions

Radiologic characterization of pancreatic lesions is currently limited. Computed tomography is insensitive in detecting and characterizing small pancreatic lesions. Moreover, heterogeneity of many pancreatic lesions makes determination of malignancy challenging. As a result, invasive diagnostic testing is frequently used to characterize pancreatic lesions but often yields indeterminate results. Computed tomography texture analysis (CTTA) is an emerging noninvasive computational tool that quantifies gray-scale pixels/voxels and their spatial relationships within a region of interest. In nonpancreatic lesions, CTTA has shown promise in diagnosis, lesion characterization, and risk stratification, and more recently, pancreatic applications of CTTA have been explored. This review outlines the emerging role of CTTA in identifying, characterizing, and risk stratifying pancreatic lesions. Although recent studies show the clinical potential of CTTA of the pancreas, a clear understanding of which specific texture features correlate with high-grade dysplasia and predict survival has not yet been achieved. Further multidisciplinary investigations using strong radiologic-pathologic correlation are needed to establish a role for this noninvasive diagnostic tool.

Computed Tomography-Based Radiomic Features for Diagnosis of Indeterminate Small Pulmonary Nodules

OBJECTIVE

This study aimed to determine the potential of radiomic features extracted from preoperative computed tomography to discriminate malignant from benign indeterminate small (≤10 mm) pulmonary nodules.

METHODS

A total of 197 patients with 210 nodules who underwent surgical resections between January 2011 and March 2017 were analyzed. Three hundred eighty-five radiomic features were extracted from the computed tomographic images. Feature selection and data dimension reduction were performed using the Kruskal-Wallis test, Spearman correlation analysis, and principal component analysis. The random forest was used for radiomic signature building. The receiver operating characteristic curve analysis was used to evaluate the model performance.

RESULTS

Fifteen principal component features were selected for modeling. The area under the curve, sensitivity, specificity, and accuracy of the prediction model were 0.877 (95% confidence interval [CI], 0.795-0.959), 81.8% (95% CI, 72.0%-90.9%), 77.4% (95% CI, 63.9%-89.3%), and 80.0% (95% CI, 72.0%-86.7%) in the validation cohort, respectively.

CONCLUSIONS

Computed tomography-based radiomic features showed good discriminative power for benign and malignant indeterminate small pulmonary nodules.

Radiomics predicts survival of patients with advanced non-small cell lung cancer undergoing PD-1 blockade using Nivolumab

Immune checkpoint blockade is an emerging anticancer strategy, and Nivolumab is a human mAb to PD-1 that is used in the treatment of a number of different malignancies, including non-small cell lung cancer (NSCLC), kidney cancer, urothelial carcinoma and melanoma. Although the use of Nivolumab prolongs survival in a number of patients, this treatment is hampered by high cost. Therefore, the identification of predictive markers of response to treatment in patients is required. In this context, PD-1/PDL1 blockade antitumor effects occur through the reactivation of a pre-existing immune response, and the efficacy of these effects is strictly associated with the presence of necrosis, hypoxia and inflammation at the tumour sites. It has been indicated that these events can be evaluated by specific assessments using a computed tomography (CT) texture analysis (TA) or radiomics. Therefore, a retrospective study was performed, which aimed to evaluate the potential use of this analysis in the identification of patients with NSCLC who may benefit from Nivolumab treatment. A retrospective analysis was performed of 59 patients with metastatic NSCLC who received Nivolumab treatment between January 2015 and July 2017 at Siena University Hospital (35 patients, training dataset), Catanzaro University Hospital and Reggio Calabria Grand Metropolitan Hospital, Italy (24 patients, validation dataset). Pre- and post-contrast CT sequences were used to contour the gross tumour volume (GTV) of the target lesions prior to Nivolumab treatment. The impact of variations on contouring was analysed using two delineations, which were performed on each patient, and the TA parameters were tested for reliability using the Intraclass Coefficient Correlation method (ICC). All analyses for the current study were performed using LifeX Software^©. Imaging, clinical and pathological parameters were correlated with progression free survival and overall survival (OS) using Kaplan Meier analysis. An external validation testing was performed for the TA Score using the validation dataset. A total of 59 patients were included in the analysis of the present study. The reliability ICC analysis of 14 TA parameters indicated a highly reproducibility (ICC >0.70, single measure) in 12 (85%) pre- contrast and 13 (93%) post-contrast exams. A specific cut-off was detected for each of the following parameters: volume (score 1 >36 ml), histogram entropy (score 1 > 1.30), compacity (score 1 <3), gray level co-occurrence matrix (GLCM)-entropy (score 1 >1.80), GLCM-Dissimilarity (score 1 >5) and GLCM-Correlation (score 1<0.54). The global texture score allowed the classification of two subgroups of Low (Score 0-1; 36 patients; 61%) and High Risk patients (Score >1; 23 patients; 39%) that respectively, showed a median OS of 26 (mean +/- SD: 18 +/- 1.98 months; 95% CI 14-21 months) and 5 months (mean +/- SD: 6 +/- 0.99 months; 95% CI: 4-8 months; P=0.002). The current study indicated that TA parameters can identify patients that will benefit from PD-1 blockage by defining the radiological settings that are potentially suggestive of an active immune response. These results require further confirmation in prospective trials.

Pancreatic Ductal Adenocarcinoma: Machine Learning-Based Quantitative Computed Tomography Texture Analysis For Prediction Of Histopathological Grade

Purpose

To assess the performance of combining computed tomography (CT) texture analysis with machine learning for discriminating different histopathological grades of pancreatic ductal adenocarcinoma (PDAC).

Methods

From July 2012 to August 2017, this retrospective study comprised 56 patients with confirmed histopathological PDAC (32 men, 24 women, mean age 64.04±7.82 years) who had undergone preoperative contrast-enhanced CT imaging within 1 month before surgery. Two radiologists blinded to the histopathological outcome independently segmented lesions for quantitative texture analysis. Histogram features, co-occurrence, and run-length texture were calculated. A support-vector machine was constructed to predict the pathological grade of PDAC based on preoperative texture features.

Results

Pathological analysis confirmed 37 low-grade PDAC (five well-differentiated/grade I and 32 moderately differentiated/grade II) and 19 high-grade PDAC (19 poorly differentiated/grade III) tumors. There were no significant differences in clinical or biological characteristics between patients with high-grade and low-grade tumors (P>0.05). There were significant differences between low-grade PDAC and high-grade PDAC on nine histogram features, seven run-length features, and two co-occurrence features. Cluster shade was the most important predictor (sensitivity 0.315). Using these texture features, the support-vector machine achieved 86% accuracy, 78% sensitivity, 95% and specificity.

Conclusion

Machine learning-based CT texture analysis accurately predicted histopathological differentiation grade of PDAC based on preoperative texture features, leading to maximization patient survival and achievement of personalized precision treatment.

Evaluating the tumor biology of lung adenocarcinoma: A multimodal analysis

We evaluated the relationships among functional imaging modality such as PET-CT and DW-MRI and lung adenocarcinoma pathologic heterogeneity, extent of invasion depth, and tumor cellularity as a marker of tumor microenvironment.In total, 74 lung adenocarcinomas were prospectively included. All patients underwent 18F-fluorodeoxyglucose (FDG) PET-CT and MRI before curative surgery. Pathology revealed 68 stage I tumors, 3 stage II tumors, and 3 stage IIIA tumors. Comprehensive histologic subtyping was performed for all surgically resected tumors. Maximum standardized uptake value (SUVmax) and ADC values were correlated with pathologic grade, extent of invasion, solid tumor size, and tumor cellularity.Mean solid tumor size (low: 1.7 ± 3.0 mm, indeterminate: 13.9 ± 14.2 mm, and high grade: 30.3 ± 13.5 mm) and SUVmax (low: 1.5 ± 0.2, indeterminate: 3.5 ± 2.5, and high grade: 15.3 ± 0) had a significant relationship with pathologic grade based on 95% confidence intervals (P = .01 and P < .01, respectively). SUVmax showed a strong correlation with tumor cellularity (R = 0.713, P < .001), but was not correlated with extent of invasion (R = 0.387, P = .148). A significant and strong positive correlation was observed among SUVmax values and higher cellularity and pathologic grade. ADC did not exhibit a significant relationship with tumor cellularity.Intratumor heterogeneity quantification using a multimodal-multiparametric approach might be effective when tumor volume consists of a real tumor component as well as a non-tumorous stromal component.

CT Texture Analysis-Correlations With Histopathology Parameters in Head and Neck Squamous Cell Carcinomas

Introduction: Texture analysis is an emergent imaging technique to quantify heterogeneity in radiological images. It is still unclear whether this technique is capable to reflect tumor microstructure. The present study sought to correlate histopathology parameters with texture features derived from contrast-enhanced CT images in head and neck squamous cell carcinomas (HNSCC).

Materials and Methods: Twenty-eight patients with histopathological proven HNSCC were retrospectively analyzed. In every case EGFR, VEGF, Hif1-alpha, Ki67, p53 expression derived from immunhistochemical specimen were semiautomatically calculated. Furthermore, mean cell count was estimated. Texture analysis was performed on contrast-enhanced CT images as a whole lesion measurement. Spearman's correlation analysis was performed, adjusted with Benjamini-Hochberg correction for multiple tests.

Results: Several texture features correlated with histopathological parameters. After correction only CT texture joint entropy and CT entropy correlation with Hif1-alpha expression remained statistically significant (ρ = −0.60 and ρ = −0.50, respectively).

Conclusions: CT texture joint entropy and CT entropy were associated with Hif1-alpha expression in HNSCC and might be able to reflect hypoxic areas in this entity.

Prediction of ISUP grading of clear cell renal cell carcinoma using support vector machine model based on CT images

BACKGROUND

To explore whether radiomics combined with computed tomography (CT) images can be used to establish a model for differentiating high grade (International Society of Urological Pathology [ISUP] grade III-IV) from low-grade (ISUP I-II) clear cell renal cell carcinoma (ccRCC).

METHODS

For this retrospective study, 3-phase contrast-enhanced CT images were collected from 227 patients with pathologically confirmed ISUP-grade ccRCC (155 cases in the low-grade group and 72 cases in the high-grade group). First, we delineated the largest dimension of the tumor in the corticomedullary and nephrographic CT images to obtain the region of interest. Second, variance selection, single variable selection, and the least absolute shrinkage and selection operator were used to select features in the corticomedullary phase, nephrographic phase, and 2-phase union samples, respectively. Finally, a model was constructed using the optimal features, and the receiver operating characteristic curve and area under the curve (AUC) were used to evaluate the predictive performance of the features in the training and validation queues. A Z test was employed to compare the differences in AUC values.

RESULTS

The support vector machine (SVM) model constructed using the screening features for the 2-stage joint samples can effectively distinguish between high- and low-grade ccRCC, and obtained the highest prediction accuracy. Its AUC values in the training queue and the validation queue were 0.88 and 0.91, respectively. The results of the Z test showed that the differences between the 3 groups were not statistically significant.

CONCLUSION

The SVM model constructed by CT-based radiomic features can effectively identify the ISUP grades of ccRCC.

Predicting the Development of Normal-Appearing White Matter With Radiomics in the Aging Brain: A Longitudinal Clinical Study

Background: Normal-appearing white matter (NAWM) refers to the normal, yet diseased tissue around the white matter hyperintensities (WMH) on conventional MR images. Radiomics is an emerging quantitative imaging technique that provides more details than a traditional visual analysis. This study aims to explore whether WMH could be predicted during the early stages of NAWM, using a textural analysis in the general elderly population.

Methods: Imaging data were obtained from PACS between 2012 and 2017. The subjects (≥60 years) received two or more MRI exams on the same scanner with time intervals of more than 1 year. By comparing the baseline and follow-up images, patients with noted progression of WMH were included as the case group (n = 51), while age-matched subjects without WMH were included as the control group (n = 51). Segmentations of the regions of interest (ROIs) were done with the ITK software. Two ROIs of developing NAWM (dNAWM) and non-developing NAWM (non-dNAWM) were drawn separately on the FLAIR images of each patient. dNAWM appeared normal on the baseline images, yet evolved into WMH on the follow-up images. Non-dNAWM appeared normal on both the baseline and follow-up images. A third ROI of normal white matter (NWM) was extracted from the control group, which was normal on both baseline and follow-up images. Textural features were dimensionally reduced with ANOVA+MW, correlation analysis, and LASSO. Three models were built based on the optimal parameters of dimensional reduction, including Model 1 (NWM vs. dNAWM), Model 2 (non-dNAWM vs. dNAWM), and Model 3 (NWM vs. non-dNAWM). The ROC curve was adopted to evaluate the classification validity of these models.

Results: Basic characteristics of the patients and controls showed no significant differences. The AUC of Model 1 in training and test groups were 0.967 (95% CI: 0.831–0.999) and 0.954 (95% CI: 0.876–0.989), respectively. The AUC of Model 2 were 0.939 (95% CI: 0.856–0.982) and 0.846 (95% CI: 0.671–0.950). The AUC of Model 3 were 0.713 (95% CI: 0.593–0.814) and 0.667 (95% CI: 0.475–0.825).

Conclusion: Radiomics textural analysis can distinguish dNAWM from non-dNAWM on FLAIR images, which could be used for the early detection of NAWM lesions before they develop into visible WHM.

Peritumoral radiomics features predict distant metastasis in locally advanced NSCLC

Purpose

Radiomics provides quantitative tissue heterogeneity profiling and is an exciting approach to developing imaging biomarkers in the context of precision medicine. Normal-appearing parenchymal tissues surrounding primary tumors can harbor microscopic disease that leads to increased risk of distant metastasis (DM). This study assesses whether computed-tomography (CT) imaging features of such peritumoral tissues can predict DM in locally advanced non-small cell lung cancer (NSCLC).

Material and methods

200 NSCLC patients of histological adenocarcinoma were included in this study. The investigated lung tissues were tumor rim, defined to be 3mm of tumor and parenchymal tissue on either side of the tumor border and the exterior region extended from 3 to 9mm outside of the tumor. Fifteen stable radiomic features were extracted and evaluated from each of these regions on pre-treatment CT images. For comparison, features from expert-delineated tumor contours were similarly prepared. The patient cohort was separated into training and validation datasets for prognostic power evaluation. Both univariable and multivariable analyses were performed for each region using concordance index (CI).

Results

Univariable analysis reveals that six out of fifteen tumor rim features were significantly prognostic of DM (p-value < 0.05), as were ten features from the visible tumor, and only one of the exterior features was. Multivariablely, a rim radiomic signature achieved the highest prognostic performance in the independent validation sub-cohort (CI = 0.64, p-value = 2.4×10⁻⁵) significantly over a multivariable clinical model (CI = 0.53), a visible tumor radiomics model (CI = 0.59), or an exterior tissue model (CI = 0.55). Furthermore, patient stratification by the combined rim signature and clinical predictor led to a significant improvement on the clinical predictor alone and also outperformed stratification using the combined tumor signature and clinical predictor.

Conclusions

We identified peritumoral rim radiomic features significantly associated with DM. This study demonstrated that peritumoral imaging characteristics may provide additional valuable information over the visible tumor features for patient risk stratification due to cancer metastasis.

Data Analysis Strategies in Medical Imaging

Radiographic imaging continues to be one of the most effective and clinically useful tools within oncology. Sophistication of artificial intelligence has allowed for detailed quantification of radiographic characteristics of tissues using predefined engineered algorithms or deep learning methods. Precedents in radiology as well as a wealth of research studies hint at the clinical relevance of these characteristics. However, critical challenges are associated with the analysis of medical imaging data. Although some of these challenges are specific to the imaging field, many others like reproducibility and batch effects are generic and have already been addressed in other quantitative fields such as genomics. Here, we identify these pitfalls and provide recommendations for analysis strategies of medical imaging data, including data normalization, development of robust models, and rigorous statistical analyses. Adhering to these recommendations will not only improve analysis quality but also enhance precision medicine by allowing better integration of imaging data with other biomedical data sources. Clin Cancer Res; 24(15); 3492-9. ©2018 AACR.

Prognostic value of radiomic analysis of iodine overlay maps from dual-energy computed tomography in patients with resectable lung cancer

OBJECTIVES

To investigate whether radiomics on iodine overlay maps from dual-energy computed tomography (DECT) can predict survival outcomes in patients with resectable lung cancer.

METHODS

Ninety-three lung cancer patients eligible for curative surgery were examined with DECT at the time of diagnosis. The median follow-up was 60.4 months. Radiomic features of the entire primary tumour were extracted from iodine overlay maps generated by DECT. A Cox proportional hazards regression model was used to determine independent predictors of overall survival (OS) and disease-free survival (DFS), respectively.

RESULTS

Forty-two patients (45.2%) had disease recurrence and 39 patients (41.9%) died during the follow-up period. The mean DFS was 49.8 months and OS was 55.2 months. Univariate analysis revealed that significant predictors of both OS and DFS were stage and radiomic parameters, including histogram energy, histogram entropy, grey-level co-occurrence matrix (GLCM) angular second moment, GLCM entropy and homogeneity. The multivariate analysis identified stage and entropy as independent risk factors predicting both OS (stage, hazard ratio (HR) = 2.020 [95% CI 1.014-4.026], p = 0.046; entropy, HR = 1.543 [95% CI 1.069-2.228], p = 0.021) and DFS (stage, HR = 2.132 [95% CI 1.060-4.287], p = 0.034; entropy, HR = 1.497 [95% CI 1.031-2.173], p = 0.034). The C-index showed that adding entropy improved prediction of OS compared to stage only (0.720 and 0.667, respectively; p = 0.048).

CONCLUSIONS

Radiomic features extracted from iodine overlay map reflecting heterogeneity of tumour perfusion can add prognostic information for patients with resectable lung cancer.

KEY POINTS

• Radiomic feature (histogram entropy) from DECT iodine overlay maps was an independent risk factor predicting both overall survival and disease-free survival. • Adding histogram entropy to clinical stage improved prediction of overall survival compared to stage only (0.720 and 0.667, respectively; p = 0.048). • DECT can be a good option for comprehensive pre-operative evaluation in cases of resectable lung cancer.

Predicting survival time of lung cancer patients using radiomic analysis

OBJECTIVES

This study investigates the prediction of Non-small cell lung cancer (NSCLC) patient survival outcomes based on radiomic texture and shape features automatically extracted from tumor image data.

MATERIALS AND METHODS

Retrospective analysis involves CT scans of 315 NSCLC patients from The Cancer Imaging Archive (TCIA). A total of 24 image features are computed from labeled tumor volumes of patients within groups defined using NSCLC subtype and TNM staging information. Spearman's rank correlation, Kaplan-Meier estimation and log-rank tests were used to identify features related to long/short NSCLC patient survival groups. Automatic random forest classification was used to predict patient survival group from multivariate feature data. Significance is assessed at P < 0.05 following Holm-Bonferroni correction for multiple comparisons.

RESULTS

Significant correlations between radiomic features and survival were observed for four clinical groups: (group, [absolute correlation range]): (large cell carcinoma (LCC) [0.35, 0.43]), (tumor size T2, [0.31, 0.39]), (non lymph node metastasis N0, [0.3, 0.33]), (TNM stage I, [0.39, 0.48]). Significant log-rank relationships between features and survival time were observed for three clinical groups: (group, hazard ratio): (LCC, 3.0), (LCC, 3.9), (T2, 2.5) and (stage I, 2.9). Automatic survival prediction performance (i.e. below/above median) is superior for combined radiomic features with age-TNM in comparison to standard TNM clinical staging information (clinical group, mean area-under-the-ROC-curve (AUC)): (LCC, 75.73%), (N0, 70.33%), (T2, 70.28%) and (TNM-I, 76.17%).

CONCLUSION

Quantitative lung CT imaging features can be used as indicators of survival, in particular for patients with large-cell-carcinoma (LCC), primary-tumor-sizes (T2) and no lymph-node-metastasis (N0).

Long Noncoding RNA CRNDE/PRC2 Participated in the Radiotherapy Resistance of Human Lung Adenocarcinoma Through Targeting p21 Expression

Long noncoding RNAs (lncRNAs), a new class of functional regulators involved in human tumorigenesis, have been attracting the increasing attention of researchers. The lncRNA colorectal neoplasia differentially expressed (CRNDE) gene, transcribed from chromosome 16 on the strand opposite the adjacent IRX5 gene, was originally found to be increased in CRC and was reported to be abnormally expressed in many cancers. However, its potential role and the molecular mechanism underlying the radioresistant phenotype formation of lung adenocarcinoma (LAD) remain unclear. In our present study, we identified that CRNDE was significantly upregulated in LAD tissue and radioresistant LAD cell lines. A high level of CRNDE expression was significantly correlated with poor differentiation, TNM stage, lymph node metastasis, radiotherapy response, and a significantly shorter overall survival. Gain- and loss-of-function tests revealed that CRNDE could influence the radiosensitivity of LAD cells by affecting the G₁/S transition and causing apoptosis of LAD cells in vitro. Additionally, the mechanistic investigations showed that CRNDE could interact with PRC2 and recruit its core component EZH2 to p21 (CDKN1A) promoter regions and repress its transcription. Furthermore, rescue experiments were performed to confirm that CRNDE oncogenic function was partly through regulating p21. In conclusion, our data suggest that CRNDE may function as an oncogene by modulating p21, finally contributing to the radioresistant phenotype formation of LAD cells.