Search

Objective. Tracer kinetic models allow for estimating pharmacokinetic (PK) parameters, which are related to pathological characteristics, from breast dynamic contrast-enhanced magnetic resonance imaging. However, existing tracer kinetic models subject to inaccuracy are time-consuming for PK parameters estimation. This study aimed to accurately and efficiently estimate PK parameters for predicting molecular subtypes based on convolutional neural network (CNN). Approach. A CNN integrating global and local features (GL-CNN) was trained using synthetic data where known PK parameters map was used as the ground truth, and subsequently used to directly estimate PK parameters (volume transfer constant K trans and flux rate constant K ep) map. The accuracy assessed by the peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and concordance correlation coefficient (CCC) was compared between the GL-CNN and Tofts-based PK parameters in synthetic data. Radiomic features were calculated from the PK parameters map in 208 breast tumors. A random forest classifier was constructed to predict molecular subtypes using a discovery cohort (n = 144). The diagnostic performance evaluated on a validation cohort (n = 64) using the area under the receiver operating characteristic curve (AUC) was compared between the GL-CNN and Tofts-based PK parameters. Main results. The average PSNR (48.8884), SSIM (0.9995), and CCC (0.9995) between the GL-CNN-based K trans map and ground truth were significantly higher than those between the Tofts-based K trans map and ground truth. The GL-CNN-based K trans obtained significantly better diagnostic performance (AUCs = 0.7658 and 0.8528) than the Tofts-based K trans for luminal B and HER2 tumors. The GL-CNN method accelerated the computation by speed approximately 79 times compared to the Tofts method for the whole breast of all patients. Significance. Our results indicate that the GL-CNN method can be used to accurately and efficiently estimate PK parameters for predicting molecular subtypes. [ABSTRACT FROM AUTHOR]

Aiming at accurate survival prediction of Glioblastoma (GBM) patients following radiation therapy, we developed a subregion-based survival prediction framework via a novel feature construction method on multi-sequence MRIs. The proposed method consists of two main steps: (1) a feature space optimization algorithm to determine the most appropriate matching relation derived between multi-sequence MRIs and tumor subregions, for using multimodal image data more reasonable; (2) a clustering-based feature bundling and construction algorithm to compress the high-dimensional extracted radiomic features and construct a smaller but effective set of features, for accurate prediction model construction. For each tumor subregion, a total of 680 radiomic features were extracted from one MRI sequence using Pyradiomics. Additional 71 geometric features and clinical information were collected resulting in an extreme high-dimensional feature space of 8231 to train and evaluate the survival prediction at 1 year, and the more challenging overall survival prediction. The framework was developed based on 98 GBM patients from the BraTS 2020 dataset under five-fold cross-validation, and tested on an external cohort of 19 GBM patients randomly selected from the same dataset. Finally, we identified the best matching relationship between each subregion and its corresponding MRI sequence, a subset of 235 features (out of 8231 features) were generated by the proposed feature bundling and construction framework. The subregion-based survival prediction framework achieved AUCs of 0.998 and 0.983 on the training and independent test cohort respectively for 1 year survival prediction, compared to AUCs of 0.940 and 0.923 for survival prediction using the 8231 initial extracted features for training and validation cohorts respectively. Finally, we further constructed an effective stacking structure ensemble regressor to predict the overall survival with the C-index of 0.872. The proposed subregion-based survival prediction framework allow us to better stratified patients towards personalized treatment of GBM. [ABSTRACT FROM AUTHOR]

Objective. To propose a new dynamic multiparametric magnetic resonance imaging (mpMRI) radiomics method for the detection of prostate cancer (PCa), and establish a combined model using dynamic and static radiomics features based on this method. Approach. A total of 166 patients (82 PCa patients and 84 non-PCa patients) were enrolled in the study, and 31 872 mpMRI images were performed in a radiomics workflow. The whole prostate segmentation and traditional static radiomics features extraction were performed on intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI, b value of 10, 50, 100, 150, 200, 400, 600, 800, 1000, 1500 s mm−2 respectively), apparent diffusion coefficient (ADC), and T2-weighted imaging (T2WI) sequences respectively. Through the building of each b -value DWI model and the analysis of the static key radiomics features, three types of dynamic features called standard discrete (SD), parameter (P) and relative change rate (RCR) were constructed. And the b-value parameters used to construct dynamic features were divided into three groups ('Df_', 'Db_' and 'Da_'): the front part (10–200 s mm−2), the back part (400–1500 s mm−2), and all (10–1500 s mm−2) of the b -values set, respectively. Afterwards, the dynamic mpMRI model and combined model construction were constructed, and the PCa discrimination performance of each model was evaluated. Main results. The models based on dynamic features showed good potential for PCa identification. Where, the results of Db_SD, Da_P and Db_P models were encouraging (test cohort AUCs: 90.78%, 87.60%, 86.3%), which was better than the commonly used ADC model (AUC of ADC was 75.48%). Among the combined models, the models using static features of T2WI and dynamic features performed the best. The AUC of Db_SD + T2WI, Db_P + T2WI and Db_RCR + T2WI model was 92.90%, 91.29% and 81.46%. Significance. The dynamic-static combination model based on dynamic mpMRI radiomics method has a good effect on the identification of PCa. This method has broad application prospects in PCa individual diagnosis management. [ABSTRACT FROM AUTHOR]

Objective. With the progress of artificial intelligence (AI) in magnetic resonance imaging (MRI), large-scale multi-center MRI datasets have a great influence on diagnosis accuracy and model performance. However, multi-center images are highly variable due to the variety of scanners or scanning parameters in use, which has a negative effect on the generality of AI-based diagnosis models. To address this problem, we propose a self-supervised harmonization (SSH) method. Approach. Mapping the style of images between centers allows harmonization without traveling phantoms to be formalized as an unpaired image-to-image translation problem between two domains. The mapping is a two-stage transform, consisting of a modified cycle generative adversarial network (cycleGAN) for style transfer and a histogram matching module for structure fidelity. The proposed algorithm is demonstrated using female pelvic MRI images from two 3 T systems and compared with three state-of-the-art methods and one conventional method. In the absence of traveling phantoms, we evaluate harmonization from three perspectives: image fidelity, ability to remove inter-center differences, and influence on the downstream model. Main results. The improved image sharpness and structure fidelity are observed using the proposed harmonization pipeline. It largely decreases the number of features with a significant difference between two systems (from 64 to 45, lower than dualGAN: 57, cycleGAN: 59, ComBat: 64, and CLAHE: 54). In the downstream cervical cancer classification, it yields an area under the receiver operating characteristic curve of 0.894 (higher than dualGAN: 0.828, cycleGAN: 0.812, ComBat: 0.685, and CLAHE: 0.770). Significance. Our SSH method yields superior generality of downstream cervical cancer classification models by significantly decreasing the difference in radiomics features, and it achieves greater image fidelity. [ABSTRACT FROM AUTHOR]

Objective. To develop and externally validate habitat-based MRI radiomics for preoperative prediction of the EGFR mutation status based on brain metastasis (BM) from primary lung adenocarcinoma (LA). Approach. We retrospectively reviewed 150 and 38 patients from hospital 1 and hospital 2 between January 2017 and December 2021 to form a primary and an external validation cohort, respectively. Radiomics features were calculated from the whole tumor (W), tumor active area (TAA) and peritumoral oedema area (POA) in the contrast-enhanced T1-weighted (T1CE) and T2-weighted (T2W) MRI image. The least absolute shrinkage and selection operator was applied to select the most important features and to develop radiomics signatures (RSs) based on W (RS-W), TAA (RS-TAA), POA (RS-POA) and in combination (RS-Com). The area under receiver operating characteristic curve (AUC) and accuracy analysis were performed to assess the performance of radiomics models. Main results. RS-TAA and RS-POA outperformed RS-W in terms of AUC, ACC and sensitivity. The multi-region combined RS-Com showed the best prediction performance in the primary validation (AUCs, RS-Com versus RS-W versus RS-TAA versus RS-POA, 0.901 versus 0.699 versus 0.812 versus 0.883) and external validation (AUCs, RS-Com versus RS-W versus RS-TAA versus RS-POA, 0.900 versus 0.637 versus 0.814 versus 0.842) cohort. Significance. The developed habitat-based radiomics models can accurately detect the EGFR mutation in patients with BM from primary LA, and may provide a preoperative basis for personal treatment planning. [ABSTRACT FROM AUTHOR]

Objective. 4D-CBCT provides phase-resolved images valuable for radiomics analysis for outcome prediction throughout treatment courses. However, 4D-CBCT suffers from streak artifacts caused by under-sampling, which severely degrades the accuracy of radiomic features. Previously we developed group-patient-trained deep learning methods to enhance the 4D-CBCT quality for radiomics analysis, which was not optimized for individual patients. In this study, a patient-specific model was developed to further improve the accuracy of 4D-CBCT based radiomics analysis for individual patients. Approach. This patient-specific model was trained with intra-patient data. Specifically, patient planning 4D-CT was augmented through image translation, rotation, and deformation to generate 305 CT volumes from 10 volumes to simulate possible patient positions during the onboard image acquisition. 72 projections were simulated from 4D-CT for each phase and were used to reconstruct 4D-CBCT using FDK back-projection algorithm. The patient-specific model was trained using these 305 paired sets of patient-specific 4D-CT and 4D-CBCT data to enhance the 4D-CBCT image to match with 4D-CT images as ground truth. For model testing, 4D-CBCT were simulated from a separate set of 4D-CT scan images acquired from the same patient and were then enhanced by this patient-specific model. Radiomics features were then extracted from the testing 4D-CT, 4D-CBCT, and enhanced 4D-CBCT image sets for comparison. The patient-specific model was tested using 4 lung-SBRT patients’ data and compared with the performance of the group-based model. The impact of model dimensionality, region of interest (ROI) selection, and loss function on the model accuracy was also investigated. Main results. Compared with a group-based model, the patient-specific training model further improved the accuracy of radiomic features, especially for features with large errors in the group-based model. For example, the 3D whole-body and ROI loss-based patient-specific model reduces the errors of the first-order median feature by 83.67%, the wavelet LLL feature maximum by 91.98%, and the wavelet HLL skewness feature by 15.0% on average for the four patients tested. In addition, the patient-specific models with different dimensionality (2D versus 3D) or loss functions (L1 versus L1 + VGG + GAN) achieved comparable results for improving the radiomics accuracy. Using whole-body or whole-body+ROI L1 loss for the model achieved better results than using the ROI L1 loss alone as the loss function. Significance. This study demonstrated that the patient-specific model is more effective than the group-based model on improving the accuracy of the 4D-CBCT radiomic features analysis, which could potentially improve the precision for outcome prediction in radiotherapy. [ABSTRACT FROM AUTHOR]

Radiomics is an active area of research in medical image analysis, however poor reproducibility of radiomics has hampered its application in clinical practice. This issue is especially prominent when radiomic features are calculated from noisy images, such as low dose computed tomography (CT) scans. In this article, we investigate the possibility of improving the reproducibility of radiomic features calculated on noisy CTs by using generative models for denoising. Our work concerns two types of generative models—encoder–decoder network (EDN) and conditional generative adversarial network (CGAN). We then compared their performance against a more traditional 'non-local means' denoising algorithm. We added noise to sinograms of full dose CTs to mimic low dose CTs with two levels of noise: low-noise CT and high-noise CT. Models were trained on high-noise CTs and used to denoise low-noise CTs without re-training. We tested the performance of our model in real data, using a dataset of same-day repeated low dose CTs in order to assess the reproducibility of radiomic features in denoised images. EDN and the CGAN achieved similar improvements on the concordance correlation coefficients (CCC) of radiomic features for low-noise images from 0.87 [95%CI, (0.833, 0.901)] to 0.92 [95%CI, (0.909, 0.935)] and for high-noise images from 0.68 [95%CI, (0.617, 0.745)] to 0.92 [95%CI, (0.909, 0.936)], respectively. The EDN and the CGAN improved the test-retest reliability of radiomic features (mean CCC increased from 0.89 [95%CI, (0.881, 0.914)] to 0.94 [95%CI, (0.927, 0.951)]) based on real low dose CTs. These results show that denoising using EDN and CGANs could be used to improve the reproducibility of radiomic features calculated from noisy CTs. Moreover, images at different noise levels can be denoised to improve the reproducibility using the above models without need for re-training, provided the noise intensity is not excessively greater that of the high-noise CTs. To the authors' knowledge, this is the first effort to improve the reproducibility of radiomic features calculated on low dose CT scans by applying generative models. [ABSTRACT FROM AUTHOR]

Purpose. 4D-CT is routine imaging for lung cancer patients treated with stereotactic body radiotherapy. No studies have investigated optimal 4D phase selection for radiomics. We aim to determine how phase data should be used to identify prognostic biomarkers for distant failure, and test whether stability assessment is required. A phase selection approach will be developed to aid studies with different 4D protocols and account for patient differences. Methods. 186 features were extracted from the tumour and peritumour on all phases for 258 patients. Feature values were selected from phase features using four methods: (A) mean across phases, (B) median across phases, (C) 50% phase, and (D) the most stable phase (closest in value to two neighbours), coined personalised selection. Four levels of stability assessment were also analysed, with inclusion of: (1) all features, (2) stable features across all phases, (3) stable features across phase and neighbour phases, and (4) features averaged over neighbour phases. Clinical-radiomics models were built for twelve combinations of feature type and assessment method. Model performance was assessed by concordance index (c-index) and fraction of new information from radiomic features. Results. The most stable phase spanned the whole range but was most often near exhale. All radiomic signatures provided new information for distant failure prediction. The personalised model had the highest c-index (0.77), and 58% of new information was provided by radiomic features when no stability assessment was performed. Conclusion. The most stable phase varies per-patient and selecting this improves model performance compared to standard methods. We advise the single most stable phase should be determined by minimising feature differences to neighbour phases. Stability assessment over all phases decreases performance by excessively removing features. Instead, averaging of neighbour phases should be used when stability is of concern. The models suggest that higher peritumoural intensity predicts distant failure. [ABSTRACT FROM AUTHOR]

Radiomics have been extensively investigated as quantitative biomarkers that can enhance the utility of imaging studies and aid the clinical decision making process. A major challenge to the clinical translation of radiomics is their variability as a result of different imaging and reconstruction protocols. In this work, we present a novel radiomics standardization framework capable of modeling and recovering the underlying radiomic feature in images that have been corrupted by the effects of spatial resolution and noise. We focus on two classes of radiomics based on pixel value distributions—i.e. histograms and gray-level co-occurrence matrices (GLCMs). We developed a model that predicts these distributions in the presence of system blur and noise, and used that model to invert these physical effects and recover the underlying distributions. Specifically, the effect of blur on histogram and GLCM is highly image-dependent, while additive noise convolves the histogram/GLCM of the noiseless image with those of the noise. The recovery method therefore consists of two deconvolution operations: the first in the image domain to remove the effect of system blur, the second in the histogram/GLCM domain to remove the effect of noise. The performance of the proposed recovery strategy was investigated using a set of texture phantoms and an emulated computed tomography imaging chain with a range of realistic blur and noise levels. The proposed method was able to obtain histogram and GLCM estimates that closely resemble the ground truth. The method performed well across imaging conditions and significantly lowered the variability associated with different imaging protocols. This improvement also translated to better classification accuracy, where recovered radiomic values result in greater separation of radiomic clusters for two different texture phantoms as compared to values derived from the original blurred and noisy images. In summary, the novel radiomics standardization framework demonstrates high potential for mitigating radiomic variability as a result of the imaging system and can potentially be integrated as a preprocessing step towards more robust and reproducible radiomic models. [ABSTRACT FROM AUTHOR]

Objectives. This study aims to develop a computer-aided diagnosis (CADx) scheme to classify between benign and malignant ground glass nodules (GGNs), and fuse deep leaning and radiomics imaging features to improve the classification performance. Methods. We first retrospectively collected 513 surgery histopathology confirmed GGNs from two centers. Among these GGNs, 100 were benign and 413 were malignant. All malignant tumors were stage I lung adenocarcinoma. To segment GGNs, we applied a deep convolutional neural network and residual architecture to train and build a 3D U-Net. Then, based on the pre-trained U-Net, we used a transfer learning approach to build a deep neural network (DNN) to classify between benign and malignant GGNs. With the GGN segmentation results generated by 3D U-Net, we also developed a CT radiomics model by adopting a series of image processing techniques, i.e. radiomics feature extraction, feature selection, synthetic minority over-sampling technique, and support vector machine classifier training/testing, etc. Finally, we applied an information fusion method to fuse the prediction scores generated by DNN based CADx model and CT-radiomics based model. To evaluate the proposed model performance, we conducted a comparison experiment by testing on an independent testing dataset. Results. Comparing with DNN model and radiomics model, our fusion model yielded a significant higher area under a receiver operating characteristic curve (AUC) value of 0.73 ± 0.06 (P < 0.01). The fusion model generated an accuracy of 75.6%, F1 score of 84.6%, weighted average F1 score of 70.3%, and Matthews correlation coefficient of 43.6%, which were higher than the DNN model and radiomics model individually. Conclusions. Our experimental results demonstrated that (1) applying a CADx scheme was feasible to diagnosis of early-stage lung adenocarcinoma, (2) deep image features and radiomics features provided complementary information in classifying benign and malignant GGNs, and (3) it was an effective way to build DNN model with limited dataset by using transfer learning. Thus, to build a robust image analysis based CADx model, one can combine different types of image features to decode the imaging phenotypes of GGN. [ABSTRACT FROM AUTHOR]

We propose a multi-view data analysis approach using radiomics and dosiomics (R&D) texture features for predicting acute-phase weight loss (WL) in lung cancer radiotherapy. Baseline weight of 388 patients who underwent intensity modulated radiation therapy (IMRT) was measured between one month prior to and one week after the start of IMRT. Weight change between one week and two months after the commencement of IMRT was analyzed, and dichotomized at 5% WL. Each patient had a planning CT and contours of gross tumor volume (GTV) and esophagus (ESO). A total of 355 features including clinical parameter (CP), GTV and ESO (GTV&ESO) dose-volume histogram (DVH), GTV radiomics, and GTV&ESO dosiomics features were extracted. R&D features were categorized as first- (L1), second- (L2), higher-order (L3) statistics, and three combined groups, L1 + L2, L2 + L3 and L1 + L2 + L3. Multi-view texture analysis was performed to identify optimal R&D input features. In the training set (194 earlier patients), feature selection was performed using Boruta algorithm followed by collinearity removal based on variance inflation factor. Machine-learning models were developed using Laplacian kernel support vector machine (lpSVM), deep neural network (DNN) and their averaged ensemble classifiers. Prediction performance was tested on an independent test set (194 more recent patients), and compared among seven different input conditions: CP-only, DVH-only, R&D-only, DVH + CP, R&D + CP, R&D + DVH and R&D + DVH + CP. Combined GTV L1 + L2 + L3 radiomics and GTV&ESO L3 dosiomics were identified as optimal input features, which achieved the best performance with an ensemble classifier (AUC = 0.710), having statistically significantly higher predictability compared with DVH and/or CP features (p < 0.05). When this performance was compared to that with full R&D-only features which reflect traditional single-view data, there was a statistically significant difference (p < 0.05). Using optimized multi-view R&D input features is beneficial for predicting early WL in lung cancer radiotherapy, leading to improved performance compared to using conventional DVH and/or CP features. [ABSTRACT FROM AUTHOR]

The objective of this study was to explore radiomics features from longitudinal diffusion-weighted MRIs (DWIs) for pathologic treatment effect prediction in patients with localized soft tissue sarcoma (STS) undergoing hypofractionated preoperative radiotherapy (RT). Thirty patients with localized STS treated with preoperative hypofractionated RT were recruited to this longitudinal imaging study. DWIs were acquired at three time points using a 0.35 T MRI-guided radiotherapy system. Treatment effect score (TES) was obtained from the post-surgery pathology as a surrogate of treatment outcome. Patients were divided into two groups based on TES. Response prediction was first performed using a support vector machine (SVM) with only mean apparent diffusion coefficient (ADC) or delta ADC to serve as the benchmark. Radiomics features were then extracted from tumor ADC maps at each of the three time points. Logistic regression and SVM were constructed to predict the TES group using features selected by univariate analysis and sequential forward selection. Classification performance using SVM with features from different time points and with or without delta radiomics were evaluated. Prediction performance using only mean ADC or delta ADC was poor (area under the curve (AUC) < 0.7). For the radiomics study using features from all time points and corresponding delta radiomics, SVM significantly outperformed logistic regression (AUC of 0.91 ± 0.05 v.s. 0.85 ± 0.06). Prediction AUC values using single or multiple time points without delta radiomics were all below 0.74. Including delta radiomics of mid- or post-treatment relative to the baseline drastically boosted the prediction. In this work, an SVM model was built to predict the TES using radiomics features from longitudinal DWI. Based on this study, we found that use of mean ADC, delta ADC, or radiomics features alone was not sufficient for response prediction, and including delta radiomics features of mid- or post-treatment relative to the baseline can optimize the prediction of TES, a pathologic and clinical endpoint. [ABSTRACT FROM AUTHOR]

This work seeks to evaluate the combatting batch effect (ComBat) harmonization algorithm's ability to reduce the variation in radiomic features arising from different imaging protocols and independently verify published results. The Gammex computed tomography (CT) electron density phantom and Quasar body phantom were imaged using 32 different chest imaging protocols. 107 radiomic features were extracted from 15 spatially varying spherical contours between 1.5 cm and 3 cm in each of the lung300 density, lung450 density, and wood inserts. The Kolmogorov–Smirnov test was used to determine significant differences in the distribution of the features and the concordance correlation coefficient (CCC) was used to measure the repeatability of the features from each protocol variation class (kVp, pitch, etc) before and after ComBat harmonization. P-values were corrected for multiple comparisons using the Benjamini–Hochberg–Yekutieli procedure. Finally, the ComBat algorithm was applied to human subject data using six different thorax imaging protocols with 135 patients. Spherical contours of un-irradiated lung (2 cm) and vertebral bone (1 cm) were used for radiomic feature extraction. ComBat harmonization reduced the percentage of features from significantly different distributions to 0%–2% or preserved 0% across all protocol variations for the lung300, lung450 and wood inserts. For the human subject data, ComBat harmonization reduced the percentage of significantly different features from 0%–59% for bone and 0%–19% for lung to 0% for both. This work verifies previously published results and demonstrates that ComBat harmonization is an effective means to harmonize radiomic features extracted from different imaging protocols to allow comparisons in large multi-institution datasets. Biological variation can be explicitly preserved by providing the ComBat algorithm with clinical or biological variables to protect. ComBat harmonization should be tested for its effect on predictive models. [ABSTRACT FROM AUTHOR]

Radiomic features achieve promising results in cancer diagnosis, treatment response prediction, and survival prediction. Our goal is to compare the handcrafted (explicitly designed) and deep learning (DL)-based radiomic features extracted from pre-treatment diffusion-weighted magnetic resonance images (DWIs) for predicting neoadjuvant chemoradiation treatment (nCRT) response in patients with locally advanced rectal cancer (LARC). 43 Patients receiving nCRT were included. All patients underwent DWIs before nCRT and total mesorectal excision surgery 6–12 weeks after completion of nCRT. Gross tumor volume (GTV) contours were drawn by an experienced radiation oncologist on DWIs. The patient-cohort was split into the responder group (n = 22) and the non-responder group (n = 21) based on the post-nCRT response assessed by postoperative pathology, MRI or colonoscopy. Handcrafted and DL-based features were extracted from the apparent diffusion coefficient (ADC) map of the DWI using conventional computer-aided diagnosis methods and a pre-trained convolution neural network, respectively. Least absolute shrinkage and selection operator (LASSO)-logistic regression models were constructed using extracted features for predicting treatment response. The model performance was evaluated with repeated 20 times stratified 4-fold cross-validation using receiver operating characteristic (ROC) curves and compared using the corrected paired t-test. The model built with handcrafted features achieved the mean area under the ROC curve (AUC) of 0.64, while the one built with DL-based features yielded the mean AUC of 0.73. The corrected paired t-test on AUC showed P-value < 0.05. DL-based features extracted from pre-treatment DWIs achieved significantly better classification performance compared with handcrafted features for predicting nCRT response in patients with LARC. [ABSTRACT FROM AUTHOR]

To predict the epidermal growth factor receptor (EGFR) mutation status in patients with lung adenocarcinoma using quantitative radiomic biomarkers and semantic features. We analyzed the computed tomography (CT) images and medical record data of 104 patients with lung adenocarcinoma who underwent surgical excision and EGFR mutation detection from 2016 to 2018 at our center. CT radiomic and semantic features that reflect the tumors’ heterogeneity and phenotype were extracted from preoperative non-enhanced CT scans. The least absolute shrinkage and selection operator method was applied to select the most distinguishable features. Three logistic regression models were built to predict the EGFR mutation status by combining the CT semantic with clinicopathological characteristics, using the radiomic features alone, and by combining the radiomic and clinicopathological features. Receiver operating characteristic (ROC) curve analysis was performed using five-fold cross-validation and the mean area under the curve (AUC) values were calculated and compared between the models to obtain the optimal model for predicting EGFR mutation. Furthermore, radiomic nomograms were constructed to demonstrate the performance of the model. In total, 1025 radiomic features were extracted and reduced to 13 features as the most important predictors to build the radiomic signature. The combined radiomic and clinicopathological features model was developed based on the radiomic signature, sex, smoking, vascular infiltration, and pathohistological type. The AUC was 0.90  ±  0.02 for the training, 0.88  ±  0.11 for the verification, and 0.894 for the test dataset. This model was superior to the other prediction models that used the combined CT semantic and clinicopathological features (AUC for the test dataset: 0.768) and radiomic features alone (AUC for the test dataset: 0.837). The prediction model built by radiomic biomarkers and clinicopathological features, including the radiomic signature, sex, smoking, vascular infiltration, and pathological type, outperformed the other two models and could effectively predict the EGFR mutation status in patients with peripheral lung adenocarcinoma. The radiomic nomogram of this model is expected to become an effective biomarker for patients with lung adenocarcinoma requiring adjuvant targeted treatment. [ABSTRACT FROM AUTHOR]

This study aims to develop a CT-based radiomic features analysis approach for diagnosis of ground-glass opacity (GGO) pulmonary nodules, and also assess whether computer-aided diagnosis (CADx) performance changes in classifying between benign and malignant nodules associated with histopathological subtypes namely, adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC), respectively. The study involves 182 histopathology-confirmed GGO nodules collected from two cancer centers. Among them, 59 are benign, 50 are AIS, 32 are MIA, and 41 are IAC nodules. Four training/testing data sets—(1) all nodules, (2) benign and AIS nodules, (3) benign and MIA nodules, (4) benign and IAC nodules—are assembled based on their histopathological subtypes. We first segment pulmonary nodules depicted in CT images by using a 3D region growing and geodesic active contour level set algorithm. Then, we computed and extracted 1117 quantitative imaging features based on the 3D segmented nodules. After conducting radiomic features normalization process, we apply a leave-one-out cross-validation (LOOCV) method to build models by embedding with a Relief feature selection, synthetic minority oversampling technique (SMOTE) and three machine-learning classifiers namely, support vector machine classifier, logistic regression classifier and Gaussian Naïve Bayes classifier. When separately using four data sets to train and test three classifiers, the average areas under receiver operating characteristic curves (AUC) are 0.75, 0.55, 0.77 and 0.93, respectively. When testing on an independent data set, our scheme yields higher accuracy than two radiologists (61.3% versus radiologist 1: 53.1% and radiologist 2: 56.3%). This study demonstrates that: (1) the feasibility of using CT-based radiomic features analysis approach to distinguish between benign and malignant GGO nodules, (2) higher performance of CADx scheme in diagnosing GGO nodules comparing with radiologist, and (3) a consistently positive trend between classification performance and invasive grade of GGO nodules. Thus, to improve the CADx performance in diagnosing of GGO nodules, one should assemble an optimal training data set dominated with more nodules associated with non-invasive lung adenocarcinoma (i.e. AIS and MIA). [ABSTRACT FROM AUTHOR]