Quantitative CT analysis of pulmonary pure ground-glass nodule predicts histological invasiveness

Evaluation of the invasiveness of pure ground-glass nodules based on dual-head ResNet technique

OBJECTIVE

To intelligently evaluate the invasiveness of pure ground-glass nodules with multiple classifications using deep learning.

METHODS

pGGNs in 1136 patients were pathologically confirmed as lung precursor lesions [atypical adenomatous hyperplasia (AAH) and adenocarcinoma in situ (AIS)], minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC). Four different models [EfficientNet-b0 2D, dual-head ResNet_3D, a 3D model combining three features (3D_3F), and a 3D model combining 19 features (3D_19F)] were constructed to evaluate the invasiveness of pGGNs using the EfficientNet and ResNet networks. The Obuchowski index was used to evaluate the differences in diagnostic efficiency among the four models.

RESULTS

The patients with pGGNs (360 men, 776 women; mean age, 54.63 ± 12.36 years) included 235 cases of AAH + AIS, 332 cases of MIA, and 569 cases of IAC. In the validation group, the areas under the curve in detecting the invasiveness of pGGNs as a three-category classification (AAH + AIS, MIA, IAC) were 0.8008, 0.8090, 0.8165, and 0.8158 for EfficientNet-b0 2D, dual-head ResNet_3D, 3D_3F, and 3D_19F, respectively, whereas the accuracies were 0.6422, 0.6158, 0.651, and 0.6364, respectively. The Obuchowski index revealed no significant differences in the diagnostic performance of the four models.

CONCLUSIONS

The dual-head ResNet_3D_3F model had the highest diagnostic efficiency for evaluating the invasiveness of pGGNs in the four models.

Maximal diameters and mean computed tomography (CT) value of synchronous multiple pure ground-glass opacities in lung adenocarcinoma are smaller

AIMS

Synchronous multiple pure ground-glass opacities (SMpGGOs) are observed more commonly. Nevertheless whether characteristics of SMpGGOs are similar to those of solitary pure ground-glass opacities (SpGGOs), remains unknown. This retrospective study aimed to compare radiographic characteristics between SMpGGOs and SpGGOs.

MATERIALS AND METHODS

We retrospectively included patients along with SpGGOs or SMpGGOs at XXX between August 2018 and June 2020. They were enrolled in two groups (SpGGOs and SMpGGOs). The clinical records, pathologic features, and radiographic manifestations of two groups were collected and compared with SPSS 21.0.

RESULTS

138 patients (58 patients with 58 SpGGOs, 80 patients with 187 SMpGGOs) were evaluated. The threshold values of maximal diameters and mean computed tomography value for adenocarcinoma were 5.5 mm (sensitivity 86.4%, specificity 55.6%, AUC 0.777) and -615.0 Hu in SMpGGOs (sensitivity 61.4%, specificity 66.7%, AUC 0.651) for SMpGGOs, whereas 12.5 mm (sensitivity 54.5%, specificity 100%, AUC 0.851) and -531.9 Hu (sensitivity 43.2%, specificity 100%, AUC 0.724) in SpGGOs.

CONCLUSION

The threshold values of maximal diameters and mean computed tomography value for adenocarcinoma in SMpGGOs may be smaller than those in SpGGOs (5.5 mm vs. 12.5mm, -615.0 Hu vs. -531.9 Hu).

Prediction of the Benign or Malignant Nature of Pulmonary Pure Ground-Glass Nodules Based on Radiomics Analysis of High-Resolution Computed Tomography Images

To evaluate the efficacy of radiomics features extracted from preoperative high-resolution computed tomography (HRCT) scans in distinguishing benign and malignant pulmonary pure ground-glass nodules (pGGNs), a retrospective study of 395 patients from 2016 to 2020 was conducted. All nodules were randomly divided into the training and validation sets in the ratio of 7:3. Radiomics features were extracted using MaZda software (version 4.6), and the least absolute shrinkage and selection operator (LASSO) was employed for feature selection. Significant differences were observed in the training set between benign and malignant pGGNs in sex, mean CT value, margin, pleural retraction, tumor-lung interface, and internal vascular change, and then the mean CT value and the morphological features model were constructed. Fourteen radiomics features were selected by LASSO for the radiomics model. The combined model was developed by integrating all selected radiographic and radiomics features using logistic regression. The AUCs in the training set were 0.606 for the mean CT value, 0.718 for morphological features, 0.756 for radiomics features, and 0.808 for the combined model. In the validation set, AUCs were 0.601, 0.692, 0.696, and 0.738, respectively. The decision curves showed that the combined model demonstrated the highest net benefit.

Quantitative analysis of lung lesions using unenhanced chest computed tomography images

INTRODUCTION

Chest radiograph and computed tomography (CT) scans can accidentally reveal pulmonary nodules. Malignant and benign pulmonary nodules can be difficult to distinguish without specific imaging features, such as calcification, necrosis, and contrast enhancement. However, these lesions may exhibit different image texture characteristics which cannot be assessed visually. Thus, a computer-assisted quantitative method like histogram analysis (HA) of Hounsfield unit (HU) values can improve diagnostic accuracy, reducing the need for invasive biopsy.

METHODS

In this exploratory control study, nonenhanced chest CT images of 20 patients with benign (10) and cancerous (10) lesion were selected retrospectively. The appearances of benign and malignant lesions were very similar in chest CT images, and only pathology report was used to discriminate them. Free hand region of interest (ROI) was inserted inside the lesion for all slices of each lesion. Mean, minimum, maximum, and standard deviations of HU values were recorded and used to make HA.

RESULTS

HA showed that the most malignant lesions have a mean HU value between 30 and 50, a maximum HU less than 150, and a minimum HU between -30 and 20. Lesions outside these ranges were mostly benign.

CONCLUSION

Quantitative CT analysis may differentiate malignant from benign lesions without specific malignancy patterns on unenhanced chest CT image.

The value of CT shape quantification in predicting pathological classification of lung adenocarcinoma

OBJECTIVE

To evaluate whether quantification of lung GGN shape is useful in predicting pathological categorization of lung adenocarcinoma and guiding the clinic.

METHODS

98 patients with primary lung adenocarcinoma were pathologically confirmed and CT was performed preoperatively, and all lesions were pathologically ≤ 30 mm in size. On CT images, we measured the maximum area of the lesion's cross-section (MA). The longest diameter of the tumor (LD) was marked with points A and B, and the perpendicular diameter (PD) was marked with points C and D, which was the longest diameter perpendicular to AB. and D, which was the longest diameter perpendicular to AB. We took angles A and B as big angle A (BiA) and small angle A (SmA). We measured the MA, LD, and PD, and for analysis we derived the LD/PD ratio and the BiA/SmA ratio. The data were analysed using the chi-square test, t-test, ROC analysis, and binary logistic regression analysis.

RESULTS

Precursor glandular lesions (PGL) and microinvasive adenocarcinoma (MIA) were distinguished from invasive adenocarcinoma (IAC) by the BiA/SmA ratio and LD, two independent factors (p = 0.007, p = 0.018). Lung adenocarcinoma pathological categorization was indicated by the BiA/SmA ratio of 1.35 and the LD of 11.56 mm with sensitivity of 81.36% and 71.79%, respectively; specificity of 71.79% and 74.36%, respectively; and AUC of 0.8357 (95% CI: 0.7558-0.9157, p < 0.001), 0.8666 (95% CI: 0.7866-0.9465, p < 0.001), respectively. In predicting the pathological categorization of lung adenocarcinoma, the area under the ROC curve of the BiA/SmA ratio combined with LD was 0.9231 (95% CI: 0.8700-0.9762, p < 0.001), with a sensitivity of 81.36% and a specificity of 89.74%.

CONCLUSIONS

Quantification of lung GGN morphology by the BiA/SmA ratio combined with LD could be helpful in predicting pathological classification of lung adenocarcinoma.

A nomogram for predicting invasiveness of lung adenocarcinoma manifesting as pure ground-glass nodules: incorporating subjective CT signs and histogram parameters based on artificial intelligence

PURPOSE

To construct a nomogram based on subjective CT signs and artificial intelligence (AI) histogram parameters to identify invasiveness of lung adenocarcinoma presenting as pure ground-glass nodules (pGGNs) and to evaluate its diagnostic performance.

METHODS

187 patients with 228 pGGNs confirmed by postoperative pathology were collected retrospectively and divided into pre-invasive group [atypical adenomatous hyperplasia (AAH) and adenocarcinoma in situ (AIS)] and invasive group [minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC)]. All pGGNs were randomly assigned to training cohort (n = 160) and validation cohort (n = 68). Nomogram was developed using subjective CT signs and AI-based histogram parameters by logistic regression analysis. The diagnostic performance was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) curve.

RESULTS

The nomogram was constructed with nodule shape, 3D mean diameter, maximum CT value, and skewness. It showed better discriminative power in differentiating invasive lesions from pre-invasive lesions with area under curve (AUC) of 0.849 (95% CI 0.790-0.909) in the training cohort and 0.831 (95% CI 0.729-0.934) in the validation cohort, which performed better than nodule shape (AUC 0.675, 95% CI 0.609-0.741), 3D mean diameter (AUC 0.762, 95% CI 0.688-0.835), maximum CT value (AUC 0.794, 95% CI 0.727-0.862), or skewness (AUC 0.594, 95% CI 0.506-0.682) alone in training cohort (for all, P < 0.05).

CONCLUSION

For pulmonary pGGNs, the nomogram based on subjective CT signs and AI histogram parameters had a good predictive ability to discriminate invasive lung adenocarcinoma from pre-invasive lung adenocarcinoma, and it has the potential to improve diagnostic efficiency and to help the patient management.

The use of the mean computed-tomography value to predict the invasiveness of ground-glass nodules: A meta-analysis

The invasiveness of ground-glass nodules (GGNs) is difficult to characterize through morphological examination. Multiple studies have independently detected a close relationship between mean computed tomography value and invasiveness of GGNs, however, their relative diagnostic accuracy is uncertain. Here, we performed a meta-analysis to validate whether the mean computed tomography value can predict the invasiveness of GGNs. Briefly, we searched the Web of Science, Embase, PubMed, Cochrane, Google Scholar, CNKI, VIP, Wanfang and SinoMed databases. The sensitivity, specificity, 95% confidence interval (CI), symmetric receiver operating characteristic curve (SROC curve) and the area under curve (AUC) were obtained using STATA 16.0 to evaluate the predictive value of the mean computed tomography value for GGNs. The presence of heterogeneity was assessed using fixed effects sensitivity analysis and I2 statistics. We used the Deek's funnel plot to evaluate the possibility of publication bias. Thirteen studies encompassing 1564 GGNs were included in our meta-analysis. Six of these studies revealed that using the mean computed tomography value for the diagnosis of pre-invasive and invasive lesions had a sensitivity and specificity of 0.75 (95% CI: 0.61-0.85) and 0.81 (95% CI: 0.74-0.86), respectively. The optimal critical value was -557 Hu. Later, eight studies were examined for the use of the mean CT value for patients with minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC); the results showed that the sensitivity was 0.78 (95% CI: 0.66-0.86) and the specificity was 0.81 (95% CI: 0.68-0.89), and the optimal critical value was -484 Hu. Therefore, the mean computed tomography value assessed via CT scan could be a significant predictor of the invasiveness of GGNs as well as a good surgical treatment guide in patients diagnosed with lung cancer. PROSPERO REGISTRATION NUMBER: CRD42020177125.

A CT-based radiomics approach to predict immediate response of radiofrequency ablation in colorectal cancer lung metastases

Objectives

To objectively and accurately assess the immediate efficacy of radiofrequency ablation (RFA) on colorectal cancer (CRC) lung metastases, the novel multimodal data fusion model based on radiomics features and clinical variables was developed.

Methods

This case-control single-center retrospective study included 479 lung metastases treated with RFA in 198 CRC patients. Clinical and radiological data before and intraoperative computed tomography (CT) scans were retrieved. The relative radiomics features were extracted from pre- and immediate post-RFA CT scans by maximum relevance and minimum redundancy algorithm (MRMRA). The Gaussian mixture model (GMM) was used to divide the data of the training dataset and testing dataset. In the process of modeling in the training set, radiomics model, clinical model and fusion model were built based on a random forest classifier. Finally, verification was carried out on an independent test dataset. The receiver operating characteristic curves (ROC) were drawn based on the obtained predicted scores, and the corresponding area under ROC curve (AUC), accuracy, sensitivity, and specificity were calculated and compared.

Results

Among the 479 pulmonary metastases, 379 had complete response (CR) ablation and 100 had incomplete response ablation. Three hundred eighty-six lesions were selected to construct a training dataset and 93 lesions to construct a testing dataset. The multivariate logistic regression analysis revealed cancer antigen 19-9 (CA19-9, p<0.001) and the location of the metastases (p< 0.05) as independent risk factors. Significant correlations were observed between complete ablation and 9 radiomics features. The best prediction performance was achieved with the proposed multimodal data fusion model integrating radiomic features and clinical variables with the highest accuracy (82.6%), AUC value (0.921), sensitivity (80.3%), and specificity (81.4%).

Conclusion

This novel multimodal data fusion model was demonstrated efficient for immediate efficacy evaluation after RFA for CRC lung metastases, which could benefit necessary complementary treatment.

Association between quantitative spectral CT parameters, Ki-67 expression, and invasiveness in lung adenocarcinoma manifesting as ground-glass nodules

BACKGROUND

Few studies about lung ground-glass nodules (GGNs) have been done using non-enhancement spectral computed tomography (CT) imaging.

PURPOSE

To examine the association between spectral CT parameters, Ki-67 expression, and invasiveness in lung adenocarcinoma manifesting as GGNs.

MATERIAL AND METHODS

Spectral CT parameters were analyzed in 106 patients with lung GGNs. The Ki-67 labeling index (Ki-67 LI) was measured, and patients were divided into low expression and high expression groups according to the number of positive-stained cells (low expression ≤10%; high expression >10%). Spectral CT parameters were compared between low and high expression groups. The correlation between spectral CT parameters and Ki-67 LI was estimated by Spearman correlation analysis. Cases were divided into a preinvasive and minimally invasive adenocarcinoma (MIA) group (atypical adenomatous hyperplasia, adenocarcinoma in situ, and MIA) and invasive adenocarcinoma (IA) group. Spectral CT parameters were compared between the two groups. The diagnostic performance was evaluated using receiver operating characteristic analysis.

RESULTS

There were significant differences in water concentration of lesions (WCL) and monochromatic CT values between the low and high expression groups. CT 40 keV had the highest correlation coefficient with Ki-67 LI. WCL and monochromatic CT values were significantly higher in the IA group than in the pre/MIA group. The value of area under the curve of CT 40 keV was 0.946 (95% confidence interval=0.905-0.988) for differentiating the two groups; the cutoff was -280.66 Hu.

CONCLUSION

Spectral CT is an effective non-invasive method for the prediction of proliferation and invasiveness in lung adenocarcinoma manifesting as GGNs.

A CT-based nomogram for predicting the risk of adenocarcinomas in patients with subsolid nodule according to the 2021 WHO classification

Purpose

To establish a nomogram for predicting the risk of adenocarcinomas in patients with subsolid nodules (SSNs) according to the 2021 WHO classification.

Methods

A total of 656 patients who underwent SSNs resection were retrospectively enrolled. Among them, 407 patients were assigned to the derivation cohort and 249 patients were assigned to the validation cohort. Univariate and multi-variate logistic regression algorithms were utilized to identity independent risk factors of adenocarcinomas. A nomogram based on the risk factors was generated to predict the risk of adenocarcinomas. The discrimination ability of the nomogram was evaluated using the concordance index (C-index), its performance was calibrated using a calibration curve, and its clinical significance was evaluated using decision curves and clinical impact curves.

Results

Lesion size, mean CT value, vascular change and lobulation were identified as independent risk factors for adenocarcinomas. The C-index of the nomogram was 0.867 (95% CI, 0.833-0.901) in derivation cohort and 0.877 (95% CI, 0.836-0.917) in validation cohort. The calibration curve showed good agreement between the predicted and actual risks. Analysis of the decision curves and clinical impact curves revealed that the nomogram had a high standardized net benefit.

Conclusions

A nomogram for predicting the risk of adenocarcinomas in patients with SSNs was established in light of the 2021 WHO classification. The developed model can be adopted as a pre-operation tool to improve the surgical management of patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40644-022-00483-1.

Fusion of CT images and clinical variables based on deep learning for predicting invasiveness risk of stage I lung adenocarcinoma

PURPOSE

To develop a novel multimodal data fusion model by incorporating computed tomography (CT) images and clinical variables based on deep learning for predicting the invasiveness risk of stage I lung adenocarcinoma that manifests as ground-glass nodules (GGNs), and compare the diagnostic performance of it with that of radiologists.

METHODS

A total of 1946 patients with solitary and histopathologically confirmed GGNs with maximum diameter less than 3 cm were retrospectively enrolled. The training dataset containing 1704 GGNs was augmented by resampling, scaling, random cropping, etc., to generate new training data. A multimodal data fusion model based on residual learning architecture and two multilayer perceptron with attention mechanism combining CT images with patient general data and serum tumor markers was built. The distance-based confidence scores (DCS) were calculated and compared among multimodal data models with different combinations. An observer study was conducted and the prediction performance of the fusion algorithms was compared with that of the two radiologists by an independent testing dataset with 242 GGNs.

RESULTS

Among the whole GGNs, 606 GGNs are confirmed as invasive adenocarcinoma (IA) and 1340 are non-IA. The proposed novel multimodal data fusion model combining CT images, patient general data and serum tumor markers achieved the highest accuracy (88.5%), Area under a ROC curve (AUC) (0.957), F1 (81.5%), F1_weighted (81.9%) and Matthews correlation coefficient (MCC) (73.2%) for classifying between IA and non-IA GGNs, which was even better than the senior radiologist's performance (accuracy, 86.1%). In addition, the DCSs for multimodal data suggested that CT image had a stronger influence (0.9540) quantitatively than general data (0.6726) or tumor marker (0.6971).

CONCLUSION

This study demonstrated that the feasibility of integrating different types of data including CT images and clinical variables, and the multimodal data fusion model yielded higher performance for distinguishing IA from non-IA GGNs. This article is protected by copyright. All rights reserved.

A tin filter's dose reduction effect revisited: Using the detectability index in low-dose computed tomography for the chest

PURPOSE

To reevaluate a tin filter's (TF) dose reduction effect in computed tomography (CT) using a combination of an anthropomorphic chest phantom and a rod-shaped phantom.

METHODS AND MATERIALS

A third-generation dual-source CT system equipped with a built-in TF was employed. A chest phantom was scanned under low-dose conditions of 0.2 to 1.0 mGy with the TF at 100 kV (TF100kV) and without it at 100 kV and 120 kV (NF100kV and NF120kV). To eliminate effects other than that of the TF, only filtered back projection (FBP) was used for image reconstruction. On the images of the rod phantom placed inside the lung field, the CT number and the spatial resolution using the modulation transfer function (MTF) were measured. Using these indices plus the noise power spectrum (NPS) that was also measured, the detectability index based on the non-prewhitening model observer (d'_NPW) was calculated.

RESULTS

The CT numbers and MTFs were almost identical across the three conditions. The area under the NPS curve was decreased by 13-17% with the TF compared with non-TF conditions. NPS increases at low frequencies of < 0.06 mm^-1 observed in NF120kV and NF100kV were eliminated by TF100kV. The potential dose reduction by the TF, estimated using the d'_NPW values, turned out to be 22 to 25%.

CONCLUSION

Based on the analysis of the FBP images of a chest phantom, the dose reduction attributable only to the TF was estimated at 22-25%, notably lower than those reported in previous studies.

Identification of pathological subtypes of early lung adenocarcinoma based on artificial intelligence parameters and CT signs

Objective: To explore the value of quantitative parameters of artificial intelligence (AI) and computed tomography (CT) signs in identifying pathological subtypes of lung adenocarcinoma appearing as ground-glass nodules (GGNs). Methods: CT images of 224 GGNs from 210 individuals were collected retrospectively and classified into atypical adenomatous hyperplasia (AAH)/adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IAC) groups. AI was used to identify GGNs and to obtain quantitative parameters, and CT signs were recognized manually. The mixed predictive model based on logistic multivariate regression was built and evaluated. Results: Of the 224 GGNs, 55, 93, and 76 were AAH/AIS, MIA, and IAC, respectively. In terms of AI parameters, from AAH/AIS to MIA, and IAC, there was a gradual increase in two-dimensional mean diameter, three-dimensional mean diameter, mean CT value, maximum CT value, and volume of GGNs (all P<0.0001). Except for the CT signs of the location, and the tumor–lung interface, there were significant differences among the three groups in the density, shape, vacuolar signs, air bronchogram, lobulation, spiculation, pleural indentation, and vascular convergence signs (all P<0.05). The areas under the curve (AUC) of predictive model 1 for identifying the AAH/AIS and MIA and model 2 for identifying MIA and IAC were 0.779 and 0.918, respectively, which were greater than the quantitative parameters independently (all P<0.05). Conclusion: AI parameters are valuable for identifying subtypes of early lung adenocarcinoma and have improved diagnostic efficacy when combined with CT signs.

Predicting the histological invasiveness of pulmonary adenocarcinoma manifesting as persistent pure ground-glass nodules by ultra-high-resolution CT target scanning in the lateral or oblique body position

Background

Ultra-high-resolution computed tomography (U-HRCT) has improved image quality for displaying the detailed characteristics of disease states and lung anatomy. The purpose of this study was to retrospectively examine whether U-HRCT target scanning in the lateral or oblique body position (protocol G scan) could predict histological invasiveness of pulmonary adenocarcinoma manifesting as pure ground-glass nodules (pGGNs).

Methods

From January 2015 to December 2016, 260 patients with 306 pathologically confirmed pGGNs who underwent preoperative protocol G scans were retrospectively reviewed and analyzed. The U-HRCT findings of preinvasive lesions [atypical adenomatous hyperplasias (AAH) and adenocarcinomas in situ (AIS)] and invasive pulmonary adenocarcinomas [minimally invasive adenocarcinomas (MIA) and invasive adenocarcinomas (IAC)] were manually compared and analyzed using orthogonal multiplanar reformation (MPR) images. The logistic regression model was established to determine variables that could predict the invasiveness of pGGNs. Receiver operating characteristic (ROC) curve analysis was performed to evaluate their diagnostic performance.

Results

There were 213 preinvasive lesions (59 AAHs and 154 AISs) and 93 invasive pulmonary adenocarcinomas (53 MIAs and 40 IACs). Compared with the preinvasive lesions, invasive adenocarcinomas exhibited a larger diameter (13.5 vs. 9.3 mm, P=0.000), higher mean attenuation (-571 vs. -613 HU, P=0.002), higher representative attenuation (-475 vs. -547 HU, P=0.000), lower relative attenuation (-339 vs. -292 HU, P=0.000) and greater frequencies of heterogeneity (P=0.001), air bronchogram (P=0.000), bubble lucency (P=0.000), and pleural indentation (P=0.000). Multiple logistic analysis revealed that larger diameter [odds ratio (OR), 1.328; 95% CI: 1.208-1.461; P=0.000] and higher representative attenuation (OR, 1.005; 95% CI: 1.003-1.007; P=0.000) were significant predictive factors of invasive pulmonary adenocarcinomas from preinvasive lesions. The optimal cut-off value of the maximum diameter for invasive pulmonary adenocarcinomas was larger than 10 mm (sensitivity, 66.7%; specificity, 72.8%).

Conclusions

The imaging features based on protocol G scanning can effectively help predict the histological invasiveness of pGGNs. The maximum diameter and representative attenuation are important parameters for predicting invasiveness.

Deep Learning-Based Stage-Wise Risk Stratification for Early Lung Adenocarcinoma in CT Images: A Multi-Center Study

Simple Summary

Prediction of the malignancy and invasiveness of ground glass nodules (GGNs) from computed tomography images is a crucial task for radiologists in risk stratification of early-stage lung adenocarcinoma. In order to solve this challenge, a two-stage deep neural network (DNN) was developed based on the images collected from four centers. A multi-reader multi-case observer study was conducted to evaluate the model capability. The performance of our model was comparable or even more accurate than that of senior radiologists, with average area under the curve values of 0.76 and 0.95 for two tasks, respectively. Findings suggest (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution reduced the model performance in predicting the risks of GGNs.

Abstract

This study aims to develop a deep neural network (DNN)-based two-stage risk stratification model for early lung adenocarcinomas in CT images, and investigate the performance compared with practicing radiologists. A total of 2393 GGNs were retrospectively collected from 2105 patients in four centers. All the pathologic results of GGNs were obtained from surgically resected specimens. A two-stage deep neural network was developed based on the 3D residual network and atrous convolution module to diagnose benign and malignant GGNs (Task1) and classify between invasive adenocarcinoma (IA) and non-IA for these malignant GGNs (Task2). A multi-reader multi-case observer study with six board-certified radiologists’ (average experience 11 years, range 2–28 years) participation was conducted to evaluate the model capability. DNN yielded area under the receiver operating characteristic curve (AUC) values of 0.76 ± 0.03 (95% confidence interval (CI): (0.69, 0.82)) and 0.96 ± 0.02 (95% CI: (0.92, 0.98)) for Task1 and Task2, which were equivalent to or higher than radiologists in the senior group with average AUC values of 0.76 and 0.95, respectively (p > 0.05). With the CT image slice thickness increasing from 1.15 mm ± 0.36 to 1.73 mm ± 0.64, DNN performance decreased 0.08 and 0.22 for the two tasks. The results demonstrated (1) a positive trend between the diagnostic performance and radiologist’s experience, (2) the DNN yielded equivalent or even higher performance in comparison with senior radiologists, and (3) low image resolution decreased model performance in predicting the risks of GGNs. Once tested prospectively in clinical practice, the DNN could have the potential to assist doctors in precision diagnosis and treatment of early lung adenocarcinoma.

High-resolution CT image analysis based on 3D convolutional neural network can enhance the classification performance of radiologists in classifying pulmonary non-solid nodules

OBJECTIVE

To investigate whether 3D convolutional neural network (CNN) is able to enhance the classification performance of radiologists in classifying pulmonary non-solid nodules (NSNs).

MATERIALS AND METHODS

Data of patients with solitary NSNs and diagnosed as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC) in pathological after surgical resection were analyzed retrospectively. Ultimately, 532 patients in our institution were included in the study: 427 cases (144 AIS, 167 MIA, 116 IAC) were assigned to training dataset and 105 cases (36 AIS, 41 MIA and 28 IAC) were assigned to validation dataset. For external validation, 177 patients (60 AIS, 69 MIA and 48 IAC) from another hospital were assigned to testing dataset. The clinical and morphological characteristics of NSNs were established as radiologists' model. The trained classification model based on 3D CNN was used to identify NSNs types automatically. The evaluation and comparison on classification performance of the two models and CNN + radiologists' model were performed via receiver operating curve (ROC) analysis and integrated discrimination improvement (IDI) index. The Akaike information criterion (AIC) was calculated to find the best-fit model.

RESULTS

In external testing dataset, radiologists' model showed inferior classification performance than CNN model both in discriminating AIS from MIA-IAC and AIS-MIA from IAC (the area under the ROC curve (Az value), 0.693 vs 0.820, P = 0.011; 0.746 vs 0.833, P = 0.026, respectively). However, combining CNN significantly enhanced the classification performance of radiologists and exhibited higher Az values than CNN model alone (Az values, 0.893 vs 0.820, P < 0.001; 0.906 vs 0.833, P < 0.001, respectively). The IDI index further confirmed CNN's contribution to radiologists in classifying NSNs (IDI = 25.8 % (18.3-46.1 %), P < 0.001; IDI = 30.1 % (26.1-45.2 %), P < 0.001, respectively). The CNN + radiologists' model also provided the best fit over radiologists' model and CNN model alone (AIC value 63.3 % vs. 29.5 %, 49.5 %, P < 0.001; 69.2 % vs. 34.9 %, 53.6 %, P < 0.001, respectively).

CONCLUSION

CNN successfully classified NSNs based on CT images and its classification performance were superior to radiologists' model. But the classification performance of radiologists can be significantly enhanced when combined with CNN in classifying NSNs.

Can peritumoral regions increase the efficiency of machine-learning prediction of pathological invasiveness in lung adenocarcinoma manifesting as ground-glass nodules?

BACKGROUND

The peri-tumor microenvironment plays an important role in the occurrence, growth and metastasis of cancer. The aim of this study is to explore the value and application of a CT image-based deep learning model of tumors and peri-tumors in predicting the invasiveness of ground-glass nodules (GGNs).

METHODS

Preoperative thin-section chest CT images were reviewed retrospectively in 622 patients with a total of 687 pulmonary GGNs. GGNs are classified according to clinical management strategies as invasive lesions (IAC) and non-invasive lesions (AAH, AIS and MIA). The two volumes of interest (VOIs) identified on CT were the gross tumor volume (GTV) and the gross volume of tumor incorporating peritumoral region (GPTV). Three dimensional (3D) DenseNet was used to model and predict GGN invasiveness, and five-fold cross validation was performed. We used GTV and GPTV as inputs for the comparison model. Prediction performance was evaluated by sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

RESULTS

The GTV-based model was able to successfully predict GGN invasiveness, with an AUC of 0.921 (95% CI, 0.896-0.937). Using GPTV, the AUC of the model increased to 0.955 (95% CI, 0.939-0.971).

CONCLUSIONS

The deep learning method performed well in predicting GGN invasiveness. The predictive ability of the GPTV-based model was more effective than that of the GTV-based model.

Can computed tomography differentiate adenocarcinoma in situ from minimally invasive adenocarcinoma?

BACKGROUND

Given the subtle pathological signs of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA), effective differentiation between the two entities is crucial. However, it is difficult to predict these conditions using preoperative computed tomography (CT) imaging. In this study, we investigated whether histological diagnosis of AIS and MIA using quantitative three-dimensional CT imaging analysis could be predicted.

METHODS

We retrospectively analyzed the images and histopathological findings of patients with lung cancer who were diagnosed with AIS or MIA between January 2017 and June 2018. We used Synapse Vincent (v. 4.3) (Fujifilm) software to analyze the CT attenuation values and performed a histogram analysis.

RESULTS

There were 22 patients with AIS and 22 with MIA. The ground-glass nodule (GGN) rate was significantly higher in patients with AIS (p < 0.001), whereas the solid volume (p < 0.001) and solid rate (p = 0.001) were significantly higher in those with MIA. The mean (p = 0.002) and maximum (p = 0.025) CT values were significantly higher in patients with MIA. The 25th, 50th, 75th, and 97.5th percentiles (all p < 0.05) for the CT values were significantly higher in patients with MIA.

CONCLUSIONS

We demonstrated that quantitative analysis of 3D-CT imaging data using software can help distinguish AIS from MIA. These analyses are useful for guiding decision-making in the surgical management of early lung cancer, as well as subsequent follow-up.

Use of relative CT values to evaluate the invasiveness of pulmonary subsolid nodules in patients with emphysema

Background

Lung cancer is a major cause of death, and adenocarcinoma is the most common histologic subtype. Precise diagnosis and treatment of invasive adenocarcinoma (IAC) can substantially improve the survival of patients. However, early-stage adenocarcinomas frequently appear as subsolid nodules (SSN) on computed tomography (CT), and the optimal cut-off CT value for differentiating the invasiveness of SSNs in emphysematous patients is unclear.

Methods

High-resolution CT targeted scans of 187 pulmonary SSNs in 175 patients with emphysema as confirmed by surgery and histology were retrospectively reviewed. The mean CT value, the relative CT (rCT) values of 1 (nodule CT value - lung CT value), and 2 (nodule CT value/lung CT value), and the size of the SSNs were measured and calculated. The differentiating performance of the CT values between pre-invasive and invasive tumors was evaluated using a receiver operating characteristic (ROC) curve.

Results

Significant differences were found in the rCT values of 1 and 2 among pure ground-glass nodules (GGNs) with different levels of invasiveness, in the rCT values of 1 and 2 for the ground-glass component (GGC) and the mean CT value of the solid component (SC) of part-solid nodules (PSNs) between minimally invasive adenocarcinoma (MIA) and IAC (all P<<0.05). The size was significantly different among pure GGNs with different invasiveness (P<0.05). The cut-off rCT values of 1, 2 and nodule size for differentiating between pre-invasive and invasive pure GGNs were 293.82 [sensitivity 58.0%, specificity 94.7%; area under the curve (AUC) 0.783], 0.68 (sensitivity 89.5%, specificity 58.0%, AUC 0.742) and 1.10 cm (sensitivity 74.0%, specificity 79.0%, AUC 0.796), respectively. The AUCs of combining rCT values 1 and 2 with the size of nodule were 0.795 (sensitivity 62.5%, specificity 89.5%) and 0.845 (sensitivity 71.6%, specificity 89.5%) respectively. There were no significant differences in the mean CT values between pure GGNs with different levels of invasiveness and between the GGC of PSNs of MIA and IAC.

Conclusions

In patients with emphysema, the rCT values are more useful than the mean CT values for differentiating between SSNs with different invasiveness and can be valuable for patient management.

The value of percentile base on computed tomography histogram in differentiating the invasiveness of adenocarcinoma appearing as pure ground-glass nodules

To investigate the value of percentile base on computed tomography (CT) histogram analysis for distinguishing invasive adenocarcinoma (IA) from adenocarcinoma in situ (AIS) or micro invasive adenocarcinoma (MIA) appearing as pure ground-glass nodules.A total of 42 cases of pure ground-glass nodules that were surgically resected and pathologically confirmed as lung adenocarcinoma between January 2015 and May 2019 were included. Cases were divided into IA and AIS/MIA in the study. The percentile on CT histogram was compared between the 2 groups. Univariate and multivariate logistic regression were used to determine which factors demonstrated a significant effect on invasiveness. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) was used to evaluate the predictive ability of individual characteristics and the combined model.The 4 histogram parameters (25th percentile, 55th percentile, 95th percentile, 97.5th percentile) and the combined model all showed a certain diagnostic value. The combined model demonstrated the best diagnostic performance. The AUC values were as follows: 25th percentile = 0.693, 55th percentile = 0.706, 95th percentile = 0.713, 97.5th percentile = 0.710, and combined model = 0.837 (all P < .05).The percentile of histogram parameters help to improve the ability to radiologically determine the invasiveness of lung adenocarcinoma appearing as pure ground-glass nodules.

How Does the Delta-Radiomics Better Differentiate Pre-Invasive GGNs From Invasive GGNs?

Purpose: This study aimed to explore the role of delta-radiomics in differentiating pre-invasive ground-glass nodules (GGNs) from invasive GGNs, compared with radiomics signature.

Materials and Methods: A total of 464 patients including 107 pre-invasive GGNs and 357 invasive GGNs were embraced in radiomics signature analysis. 3D regions of interest (ROIs) were contoured with ITK software. By means of ANOVA/MW, correlation analysis, and LASSO, the optimal radiomic features were selected. The logistic classifier of radiomics signature was constructed and radiomic scores (rad-scores) were calculated. A total of 379 patients including 48 pre-invasive GGNs and 331 invasive GGNs with baseline and follow-up CT examinations before surgeries were enrolled in delta-radiomics analysis. Finally, the logistic classifier of delta-radiomics was constructed. The receiver operating characteristic curves (ROCs) were built to evaluate the validity of classifiers.

Results: For radiomics signature analysis, six features were selected from 396 radiomic features. The areas under curve (AUCs) of logistic classifiers were 0.865 (95% CI, 0.823–0.900) in the training set and 0.800 (95% CI, 0.724–0.863) in the testing set. The rad-scores of invasive GGNs were larger than those of pre-invasive GGNs. As the follow-up interval went on, more and more delta-radiomic features became statistically different. The AUC of the delta-radiomics logistic classifier was 0.901 (95% CI, 0.867–0.928), which was higher than that of the radiomics signature.

Conclusion: The radiomics signature contributes to distinguish pre-invasive and invasive GGNs. The rad-scores of invasive GGNs were larger than those of pre-invasive GGNs. More and more delta-radiomic features appeared to be statistically different as follow-up interval prolonged. Delta-radiomics is superior to radiomics signature in differentiating pre-invasive and invasive GGNs.

Determining the invasiveness of pure ground-glass nodules using dual-energy spectral computed tomography

Background

The present work aimed to investigate the clinical application of using quantitative parameters generated in the unenhanced phase (UP) and venous phase (VP) in dual-energy spectral CT for differentiating the invasiveness of pure ground-glass nodule (pGGN).

Methods

Sixty-two patients with 66 pGGNs who underwent preoperative dual-energy spectral CT in UP and VP were evaluated retrospectively. Nodules were divided into three groups based on pathology: adenocarcinoma in situ (AIS, n=19), minimally invasive adenocarcinoma (MIA, n=22) (both in the preinvasive lesion group) and invasive adenocarcinoma (IA, n=25). The iodine concentration (IC) and water content (WC) in nodules were measured in material decomposition images. The nodule CT numbers and slopes(k) were measured on monochromatic images. All measurements, including the maximum diameter of nodules were statistically compared between the AIS-MIA group and IA group.

Results

There were significant differences of WC in VP between AIS-MIA group and IA group (P<0.05). The CT attenuation values of the 40–140 keV monochromatic images in UP and VP were significantly higher for the invasive nodules. Logistic regression analysis showed that the maximum nodule diameter [odd ratio (OR) =1.21, 95% CI: 1.050–1.400, P<0.01] and CT number in 130 keV images in venous phase (OR =1.03, 95% CI: 1.014–1.047, P<0.001) independently predicted histological invasiveness.

Conclusions

The quantitative parameters in dual-energy spectral CT in the unenhanced phase and venous phase provide useful information in differentiating preinvasive lesion group from IA group of pGGN, especially the maximum nodule diameter and CT number in the 130 keV images in the venous phase.

CT Characteristics for Predicting Invasiveness in Pulmonary Pure Ground-Glass Nodules

OBJECTIVE. The objective of our study was to investigate the differences in the CT features of atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), and invasive adenocarcinoma (IA) manifesting as a pure ground-glass nodule (pGGN) with the aim of determining parameters predictive of invasiveness. MATERIALS AND METHODS. A total of 161 patients with 172 pGGNs (14 AAHs, 59 AISs, 68 MIAs, and 31 IAs) were retrospectively enrolled. The following CT features of each histopathologic subtype of nodule were analyzed and compared: lesion location, diameter, area, shape, attenuation, uniformity of density, margin, nodule-lung interface, and internal and surrounding changes. RESULTS. ROC curves revealed that nodule diameter and area (cutoff value, 10.5 mm and 86.5 mm²; sensitivity, 87.1% and 87.1%; specificity, 70.9% and 65.2%) were significantly larger in IAs than in AAHs, AISs, and MIAs (p < 0.001), whereas the latter three were similar in size (p > 0.050). CT attenuation higher than -632 HU in pGGNs indicated invasiveness (sensitivity, 78.8%; specificity, 59.8%). As opposed to noninvasive pGGNs (AAHs and AISs), invasive pGGNs (MIAs and IAs) usually had heterogeneous density, irregular shape, coarse margin, lobulation, spiculation, pleural indentation, and dilated or distorted vessels (each, p < 0.050). Multivariate analysis showed that mean CT attenuation and presence of lobulation were predictors for invasive pGGNs (p ≤ 0.001). CONCLUSION. The likelihood of invasiveness is greater in pGGNs with larger size (> 10.5 mm or > 86.5 mm²), higher attenuation (> -632 HU), heterogeneous density, irregular shape, coarse margin, spiculation, lobulation, pleural indentation, and dilated or distorted vessels.

Comparison and Fusion of Deep Learning and Radiomics Features of Ground-Glass Nodules to Predict the Invasiveness Risk of Stage-I Lung Adenocarcinomas in CT Scan

For stage-I lung adenocarcinoma, the 5-years disease-free survival (DFS) rates of non-invasive adenocarcinoma (non-IA) is different with invasive adenocarcinoma (IA). This study aims to develop CT image based artificial intelligence (AI) schemes to classify between non-IA and IA nodules, and incorporate deep learning (DL) and radiomics features to improve the classification performance. We collect 373 surgical pathological confirmed ground-glass nodules (GGNs) from 323 patients in two centers. It involves 205 non-IA (including 107 adenocarcinoma in situ and 98 minimally invasive adenocarcinoma), and 168 IA. We first propose a recurrent residual convolutional neural network based on U-Net to segment the GGNs. Then, we build two schemes to classify between non-IA and IA namely, DL scheme and radiomics scheme, respectively. Third, to improve the classification performance, we fuse the prediction scores of two schemes by applying an information fusion method. Finally, we conduct an observer study to compare our scheme performance with two radiologists by testing on an independent dataset. Comparing with DL scheme and radiomics scheme (the area under a receiver operating characteristic curve (AUC): 0.83 ± 0.05, 0.87 ± 0.04), our new fusion scheme (AUC: 0.90 ± 0.03) significant improves the risk classification performance (p < 0.05). In a comparison with two radiologists, our new model yields higher accuracy of 80.3%. The kappa value for inter-radiologist agreement is 0.6. It demonstrates that applying AI method is an effective way to improve the invasiveness risk prediction performance of GGNs. In future, fusion of DL and radiomics features may have a potential to handle the classification task with limited dataset in medical imaging.

Radiomics for lung adenocarcinoma manifesting as pure ground-glass nodules: invasive prediction

Objectives

To investigate the value of radiomics based on CT imaging in predicting invasive adenocarcinoma manifesting as pure ground-glass nodules (pGGNs).

Methods

This study enrolled 395 pGGNs with histopathology-confirmed benign nodules or adenocarcinoma. A total of 396 radiomic features were extracted from each labeled nodule. A Rad-score was constructed with the least absolute shrinkage and selection operator (LASSO) in the training set. Multivariate logistic regression analysis was conducted to establish the radiographic model and the combined radiographic–radiomics model. The predictive performance was validated by receiver operating characteristic (ROC) curve. Based on the multivariate logistic regression analysis, an individual prediction nomogram was developed and the clinical utility was assessed.

Results

Five radiomic features and four radiographic features were selected for predicting the invasive lesions. The combined radiographic–radiomics model (AUC 0.77; 95% CI, 0.69–0.86) performed better than the radiographic model (AUC 0.71; 95% CI, 0.62–0.81) and Rad-score (AUC 0.72; 95% CI, 0.63–0.81) in the validation set. The clinical utility of the individualized prediction nomogram developed using the Rad-score, margin, spiculation, and size was confirmed in the validation set. The decision curve analysis (DCA) indicated that using a model with Rad-score to predict the invasive lesion would be more beneficial than that without Rad-score and the clinical model.

Conclusions

The proposed radiomics-based nomogram that incorporated the Rad-score, margin, spiculation, and size may be utilized as a noninvasive biomarker for the assessment of invasive prediction in patients with pGGNs.

Key Points

• CT-based radiomics analysis helps invasive prediction manifested as pGGNs.

• The combined radiographic–radiomics model may be utilized as a noninvasive biomarker for predicting invasive lesion for pGGNs.

• Radiomics-based individual nomogram may serve as a vital decision support tool to identify invasive pGGNs, obviating further workup and blind follow-up.

Electronic supplementary material

The online version of this article (10.1007/s00330-020-06776-y) contains supplementary material, which is available to authorized users.

Value of TSCT Features for Differentiating Preinvasive and Minimally Invasive Adenocarcinoma From Invasive Adenocarcinoma Presenting as Subsolid Nodules Smaller Than 3 cm

BACKGROUND

To distinguish preinvasive (adenocarcinoma in situ/atypical adenomatous hyperplasia) and minimally invasive adenocarcinoma (MIA) from invasive adenocarcinoma (IA) appearing as solitary subsolid nodules (SSNs) less than 3 cm based on thin-section computed tomography (TSCT) features to guide therapeutic approaches.

METHODS

A total of 154 lesions that were histopathologically confirmed to have pre/minimally invasive adenocarcinoma (hereafter pre/MIA) and IA presenting as part-solid nodules (PSNs) or pure ground-glass nodules (pGGNs) were retrospectively reviewed. The TSCT features, including diameter, area, CT value, shape, air bronchogram, margins, and location, were compared and assessed. Receiver operating characteristic analyses were conducted to determine the cut-off values for the qualitative variables and their diagnostic performances.

RESULTS

Of 154 nodules, 89 IA, 53 MIA, eight adenocarcinoma in situ, and four atypical adenomatous hyperplasia lesions were found. Univariate and multivariate logistic regression of the pre/MIA and IA lesions were compared and analyzed among PSNs and pGGNs. Among pGGNs, a significant difference was found in the area (p = 0.004, odds ratio [OR] = 0.124, 95% confidence interval [CI] = 0.300-0.515) between the pre/MIA and IA groups. In PSNs, significant differences were found in the diameter (p = 0.001, OR = 0.171, 95% CI = 0.063-0.467) and CT value (p = 0.001, OR = 0.996, 95% CI = 0.993-0.998) between the pre/MIA and IA groups. According to the corresponding receiver operating characteristic curves, the optimal cut-off tumor area in pGGNs to differentiate pre/MIA from IA was 0.595 cm². A higher CT value of the lesion (≥ -298.500 HU) and a larger diameter (≥1.450 cm) in PSNs were significantly associated with IA.

CONCLUSION

Imaging features from TSCT contribute to distinguishing pre/MIA from IA in solitary subsolid nodules and may contribute to guide the clinical management of these lesions.

Invasiveness and surgical timing evaluation by clinical features of ground-glass opacity nodules in lung cancers

Background

The early stages of lung cancer with ground‐glass opacity (GGO) pattern are detectable. However, it remains a challenge for physicians how best to treat GGO nodules as invasive tumors are occasionally found, even in pure GGO nodules. This study identified the invasiveness by the clinical features of the GGO nodules.

Methods

A retrospective review of patients with resected GGO nodules from August 2015 to February 2019 was performed. A total of 92 patients were enrolled and gender, age, tumor location, operation times, tumor size, histopathologic and radiological findings were analyzed.

Results

In this study, the sequential of GGO nodules invasiveness was significantly related to the tumor size and solid component. After regrouping the population into preinvasive and invasive groups, the invasiveness was significantly related to tumor size, solid component, tumor volume and maximal computed tomography (CT) value.

Conclusions

The invasiveness is difficult to evaluate according to the CT features only when the GGO nodules are less than 2 cm and consolidation/tumor ratio (C/T ratio) are less than 0.25. Tumor size and solid component are significant factors for predicting invasiveness. Part‐solid GGO nodules with a diameter greater than 1 cm require surgical consideration due to their high risk of invasiveness.

Quantitative CT Analysis for Predicting the Behavior of Part-Solid Nodules with Solid Components Less than 6 mm: Size, Density and Shape Descriptors

Persistent part-solid nodules (PSNs) with a solid component <6 mm usually represent minimally invasive adenocarcinomas and are significantly less aggressive than PSNs with a solid component ≥6 mm. However, not all PSNs with a small solid component behave in the same way: some nodules exhibit an indolent course, whereas others exhibit more aggressive behavior. Thus, predicting the future behavior of this subtype of PSN remains a complex and fascinating diagnostic challenge. The main purpose of this study was to apply open-source software to investigate which quantitative computed tomography (CT) features may be useful for predicting the behavior of a select group of PSNs. We retrospectively selected 50 patients with a single PSN with a solid component <6 mm and diameter <15 mm. Computerized analysis was performed using ImageJ software for each PSN and various quantitative features were calculated from the baseline CT images. The area, perimeter, mean Feret diameter, linear mass density, circularity and solidity were significantly related to nodule growth (p ≤ 0.031). Therefore, quantitative CT analysis was helpful for predicting the future behavior of a select group of PSNs with a solid component <6 mm and diameter <15 mm.

Computer-aided diagnosis of ground-glass opacity pulmonary nodules using radiomic features analysis

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).

Invasive Pulmonary Adenocarcinomas Versus Preinvasive Lesions Appearing as Pure Ground-Glass Nodules: Differentiation Using Enhanced Dual-Source Dual-Energy CT

OBJECTIVE. The objective of our study was to investigate the potentials of enhanced dual-source dual-energy CT (DECT) and three-planar measurements for differentiating invasive pulmonary adenocarcinomas (IPAs) from preinvasive lesions appearing as pure ground-glass nodules (pGGNs). MATERIALS AND METHODS. Thirty-nine patients with 53 pGGNs who underwent enhanced dual-source DECT were included in this retrospective study. All pGGNs were pathologically confirmed and categorized into two groups: preinvasive lesions or IPAs. The traditional CT features of the pGGNs were evaluated on unenhanced images. Quantitative parameters were measured on iodine-enhanced images of dual-source DECT in three planes, and both intra- and interobserver reproducibility analyses were performed to assess the measurement reproducibility of quantitative parameters. To identify significant factors for differentiating IPAs from preinvasive lesions, we performed logistic regression analysis and ROC curve analysis. RESULTS. For traditional CT features, only lesion size and unenhanced CT attenuation value showed significant differences between preinvasive lesions and IPAs (p < 0.05). Preinvasive lesions and IPAs exhibited significant differences in attenuation on virtual images, so-called "virtual HU" or "VHU," and the modified normalized iodine concentration (NIC) (p < 0.05), and both intra- and interobserver agreement for the quantitative measurements were excellent. Multivariate logistic regression analysis revealed that larger lesion size (adjusted odds ratio [OR], 3.65) and higher modified NIC (adjusted OR, 19.01) were significant differentiators of IPAs from preinvasive lesions (p < 0.05). ROC curve analysis revealed that modified NIC showed excellent performance (AUC, 0.924) and significantly higher performance than lesion size (AUC, 0.711) for differentiating IPAs from preinvasive lesions. CONCLUSION. In pGGNs, a lesion with a modified NIC value of more than 0.29 can be a very specific discriminator of IPAs from preinvasive lesions, and IPAs can be accurately and reliably differentiated from preinvasive lesions using enhanced dual-source DECT and three-planar measurements.

Histogram-based models on non-thin section chest CT predict invasiveness of primary lung adenocarcinoma subsolid nodules

109 pathologically proven subsolid nodules (SSN) were segmented by 2 readers on non-thin section chest CT with a lung nodule analysis software followed by extraction of CT attenuation histogram and geometric features. Functional data analysis of histograms provided data driven features (FPC1,2,3) used in further model building. Nodules were classified as pre-invasive (P1, atypical adenomatous hyperplasia and adenocarcinoma in situ), minimally invasive (P2) and invasive adenocarcinomas (P3). P1 and P2 were grouped together (T1) versus P3 (T2). Various combinations of features were compared in predictive models for binary nodule classification (T1/T2), using multiple logistic regression and non-linear classifiers. Area under ROC curve (AUC) was used as diagnostic performance criteria. Inter-reader variability was assessed using Cohen’s Kappa and intra-class coefficient (ICC). Three models predicting invasiveness of SSN were selected based on AUC. First model included 87.5 percentile of CT lesion attenuation (Q.875), interquartile range (IQR), volume and maximum/minimum diameter ratio (AUC:0.89, 95%CI:[0.75 1]). Second model included FPC1, volume and diameter ratio (AUC:0.91, 95%CI:[0.77 1]). Third model included FPC1, FPC2 and volume (AUC:0.89, 95%CI:[0.73 1]). Inter-reader variability was excellent (Kappa:0.95, ICC:0.98). Parsimonious models using histogram and geometric features differentiated invasive from minimally invasive/pre-invasive SSN with good predictive performance in non-thin section CT.

Histogram analysis combined with morphological characteristics to discriminate adenocarcinoma in situ or minimally invasive adenocarcinoma from invasive adenocarcinoma appearing as pure ground-glass nodule

OBJECTIVE

To construct a predictive model to discriminate adenocarcinoma in situ (AIS) or minimally invasive adenocarcinoma (MIA) from invasive adenocarcinoma (IAC) appearing as pure ground-glass nodules (pGGNs) using computed tomography (CT) histogram analysis combined with morphological characteristics and to evaluate its diagnostic performance.

MATERIALS AND METHODS

Two hundred eighty-nine patients with surgically resected solitary pGGN and pathologically diagnosed with AIS, MIA, or IAC in our institution from January 2014 to May 2018 were enrolled in our study. Two hundred twenty-six pGGNs (79 AIS, 84 MIA, and 63 IAC) were randomly selected and assigned to a model-development cohort, and the remaining 63 pGGNs (11 AIS, 29 MIA and 23 IAC) were assigned to a validation cohort. The morphological characteristics were established as model A and histogram parameters as model B. The diagnostic performances of model A, model B, and model A + B were evaluated and compared via receiver operating curve (ROC) analysis and logistic regression analysis.

RESULTS

Entropy (odd ratio [OR] = 23.25, 95%CI: 6.83-79.15, p < 0.001), microvascular sign (OR = 8.62, 95%CI: 3.72-19.98, p < 0.001) and the maximum diameter (OR = 4.37, 95%CI: 2.44-7.84, p < 0.001) were identified as independent predictors in the IAC group. The area under the ROC (Az value), accuracy, sensitivity and specificity of model A + B were 0.896, 88.1%, 79.4% and 91.4%, respectively, exhibiting a significantly higher Az value than either model A or model B alone (0.785 vs 0.896, p < 0.001; 0.849 vs 0.896, p = 0.029). Model A + B also conveyed a good diagnostic performance in the validation cohort, with an Az value of 0.851.

CONCLUSION

Histogram analysis combined with morphological characteristics exhibit a superior diagnostic performance in discriminating AIS-MIA from IAC appearing as pGGNs.

The "solid" component within subsolid nodules: imaging definition, display, and correlation with invasiveness of lung adenocarcinoma, a comparison of CT histograms and subjective evaluation

OBJECTIVE

To validate three proposed definitions of the "solid" component of subsolid nodules, as compared to CT histograms and the use of different window settings, for discriminating the invasiveness of adenocarcinomas in a manner that facilitates routine clinical assessment.

METHODS

We retrospectively analyzed 328 pathologically confirmed lung adenocarcinomas, manifesting as subsolid nodules. Three-dimensional CT histograms were generated by setting 11 CT attenuation intervals from - 400 to 50 HU, at 50 HU intervals, and the voxel percentage within each CT attenuation interval was generated automatically. Three definitions of the "solid" component were proposed, and 10 medium window settings were set to evaluate the "solid" component. The diagnostic performance of the three definitions for identifying invasive adenocarcinoma was compared with that of CT histogram analysis and subjective evaluation with medium window settings.

RESULTS

A parallel diagnosis using five intervals with the largest AUC (AUC ≥ 0.797) demonstrated good differential diagnostic performance, with 78% sensitivity and 73.7% specificity. Definition 2 (visibility in the mediastinum window) yielded higher accuracy (75.6%) than the other two definitions (p < 0.01). A medium window setting of - 50 WL/2 WW gave a larger AUC than the other nine medium window settings as well as definition 2, with 82.5% specificity and 88.5% PPV, which was higher than those of parallel diagnosis with CT histogram and definition 2.

CONCLUSION

Using - 50 WL/2 WW is the optimum approach for evaluating the "solid" component and discriminating invasiveness, superior to using 3D CT histograms and definition 2, and convenient in routine clinical assessment.

KEY POINTS

• - 50 WL/2 WW gave a larger AUC than definition 2. • The specificity of - 50 WL/2 WW was higher than CT histograms. • - 50 WL/2 WW offers the best evaluation of the solid component.

Hilly or mountainous surface: a new CT feature to predict the behavior of pure ground glass nodules?

•

pGGNs typically show an indolent course with very slow growth rates.

•

pGGNs exhibit different patterns of growth regardless of their initial CT features.

•

Predicting the behavior of pGGNs on initial CT remains a diagnostic challenge.

•

Diameter greater than 10 mm increases the risk of aggressive behavior in pGGNs.

•

The analysis of surface morphology may help predict the behavior of pGGNs ≥ 10 mm.

Persistent pure ground-glass nodules (pGGNs) typically show an indolent course with very slow growth rates. These slow-growing lesions exhibit different growth patterns regardless of their initial computed tomography (CT) features. Therefore, predicting the aggressive behavior of pGGNs on initial CT remains a diagnostic challenge. The literature reports that computerized analysis and various quantitative features have been tested to improve the risk stratification for pGGNs.

The present article describes the long-term follow-up of two pGGNs with different behavior and introduces, for the first time, a new computerized method of analysis that could be helpful for predicting the future behavior of pGGNs.

Validation of prediction models for risk stratification of incidentally detected pulmonary subsolid nodules: a retrospective cohort study in a Korean tertiary medical centre

OBJECTIVES

To validate the performances of two prediction models (Brock and Lee models) for the differentiation of minimally invasive adenocarcinoma (MIA) and invasive pulmonary adenocarcinoma (IPA) from preinvasive lesions among subsolid nodules (SSNs).

DESIGN

A retrospective cohort study.

SETTING

A tertiary university hospital in South Korea.

PARTICIPANTS

410 patients with 410 incidentally detected SSNs who underwent surgical resection for the pulmonary adenocarcinoma spectrum between 2011 and 2015.

PRIMARY AND SECONDARY OUTCOME MEASURES

Using clinical and radiological variables, the predicted probability of MIA/IPA was calculated from pre-existing logistic models (Brock and Lee models). Areas under the receiver operating characteristic curve (AUCs) were calculated and compared between models. Performance metrics including sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were also obtained.

RESULTS

For pure ground-glass nodules (n=101), the AUC of the Brock model in differentiating MIA/IPA (59/101) from preinvasive lesions (42/101) was 0.671. Sensitivity, specificity, accuracy, PPV and NPV based on the optimal cut-off value were 64.4%, 64.3%, 64.4%, 71.7% and 56.3%, respectively. Sensitivity, specificity, accuracy, PPV and NPV according to the Lee criteria were 76.3%, 42.9%, 62.4%, 65.2% and 56.3%, respectively. AUC was not obtained for the Lee model as a single cut-off of nodule size (≥10 mm) was suggested by this model for the assessment of pure ground-glass nodules. For part-solid nodules (n=309; 26 preinvasive lesions and 283 MIA/IPAs), the AUC was 0.746 for the Brock model and 0.771 for the Lee model (p=0.574). Sensitivity, specificity, accuracy, PPV and NPV were 82.3%, 53.8%, 79.9%, 95.1% and 21.9%, respectively, for the Brock model and 77.0%, 69.2%, 76.4%, 96.5% and 21.7%, respectively, for the Lee model.

CONCLUSIONS

The performance of prediction models for the incidentally detected SSNs in differentiating MIA/IPA from preinvasive lesions might be suboptimal. Thus, an alternative risk calculation model is required for the incidentally detected SSNs.

Effect of CT window settings on size measurements of the solid component in subsolid nodules: evaluation of prediction efficacy of the degree of pathological malignancy in lung adenocarcinoma

OBJECTIVE

To investigate the predictive value of size measurements of the solid components in pulmonary subsolid nodules with different CT window settings and to evaluate the degree of pathological malignancy in lung adenocarcinoma.  Methods: The preoperative chest CT images and pathological data of 125 patients were retrospectively evaluated. The analysis included 127 surgically resected lung adenocarcinomas that manifested as subsolid nodules. All subsolid nodules were divided into two groups: 69 in group A, including 22 adenocarcinomas in situ (AIS) and 47 minimally invasive adenocarcinomas (MIA); 58 in group B that included invasive pulmonary adenocarcinomas (IPA). The size of the solid component in the pulmonary subsolid nodules were calculated in one dimensional, two dimensional and three dimensional views using lung and mediastinal windows that were recorded as 1D-SCLW, 2D-SCLW, 3D-SCLW, 1D-SCMW, 2D-SCMW and 3D-SCMW, respectively. Furthermore, the volume of solid component with a threshold of -300HU was measured using lung window (3D-SCT). All the quantitative features were evaluated by the Mann-Whitney U test. Multivariate analysis was used to identify the significant predictor of the degree of pathological malignancy.  Results: The 1D-SCLW, 2D-SCLW, 3D-SCLW, 1D-SCMW, 2D-SCMW, 3D-SCMW and 3D-SCT views of group B were significantly larger than those of group A (p < 0.001). The multivariate logistic regression analysis indicated that 3D-SCT (OR = 1.018, 95%CI: 1.005 ~ 1.03, p <0.05＝was the independent predictive factor. The larger SCT was significantly associated with IPAs.  Conclusion: 3D-SCT of subsolid nodules during preoperative CT can be used to predict the degree of pathological malignancy in lung adenocarcinoma, which may provide a more objective and convenient selection criterion for clinical application.  Advances in knowledge:  Applying threshold of -300 HU with lung window setting would be better than other window setting for the evaluation of solid component in subsolid nodules. Computer-aided volumetry of the solid component in subsolid nodules can more accurately predict the degree of pathological malignancy than the other dimensional measurements.

Evaluation of T categories for pure ground-glass nodules with semi-automatic volumetry: is mass a better predictor of invasive part size than other volumetric parameters?

OBJECTIVES

This study aimed to investigate the diagnostic advantage of nodule mass in differentiating invasive pulmonary adenocarcinomas (IPAs) among pure ground-glass nodules (pGGNs) over other volumetric measurements. Another aim of this study was to analyse the correlation between volumetric measurements on computed tomography (CT) scans and the pathological invasive component size.

METHODS

This Institutional Review Board-approved retrospective study included 117 patients (men:women = 53:64; mean age, 57.3 years) with 117 pGGNs. Semi-automatic segmentation was performed for all nodules, and volumetric measurements, such as nodule volume, attenuation, mass, two-dimensional (2D) average diameter and three-dimensional (3D) longest diameter, were obtained. Receiver operating characteristic (ROC) curve analyses were performed to evaluate the diagnostic performances of the volumetric parameters in discriminating IPAs. Spearman correlation coefficients were calculated between the volumetric measurements and the invasive component size.

RESULTS

Area under the ROC curve for mass was 0.792 (95% CI, 0.691-0.872) in non-enhanced CT and 0.730 (95% CI, 0.607-0.832) in contrast-enhanced CT. Nodule mass was not superior to 2D average diameter for the differentiation of IPAs in both non-enhanced (0.792 vs 0.780; p = 0.501) CT and contrast-enhanced CT scans (0.730 vs 0.700; p = 0.319). The correlation between the volumetric measurements (mass, 3D longest diameter and 2D average diameter) and the invasive component size was moderate (Spearman's rho, 0.401-0.422) in non-enhanced CT and weak (Spearman's rho, 0.276-0.310) in contrast-enhanced CT.

CONCLUSIONS

Nodule mass measurement had no strength over other volumetric parameters for the prediction of pathological invasiveness in the diagnosis of pGGNs.

KEY POINTS

• Mass is not superior to other volumetric measurements for the diagnosis of pure ground-glass nodules. • Mass and two-dimensional average diameter exhibited comparable performance for the discrimination of invasive adenocarcinomas among pure ground-glass nodules. • The diagnostic performance of volumetric measurements was lower on contrast-enhanced CT scans. • The correlation between the volumetric measurements and the invasive component size was moderate on non-enhanced CT scans and weak on contrast-enhanced CT scans.

CT quantitative parameters to predict the invasiveness of lung pure ground-glass nodules (pGGNs)

AIM

To investigate the value of computed tomography (CT) quantitative parameters in predicting the invasiveness of lung pure ground-glass nodules (pGGNs).

MATERIALS AND METHODS

Chest CT images and pathological findings of 163 pGGNs in 154 consecutive patients were reviewed. According to the clinical management strategies, cases were divided into pre-invasive and MIA groups (atypical adenomatous hyperplasia [AAH], adenocarcinoma in situ [AIS], and minimally invasive adenocarcinoma [MIA]) and invasive group (invasive adenocarcinoma [IAC]). CT quantitative parameters including maximum diameter, largest diameter perpendicular to the maximum diameter, maximum cross-sectional area, volume, mass, and mean attenuation value were measured and compared between two groups. Their diagnostic performances were evaluated using receiver operating characteristic (ROC) and logistic regression analysis.

RESULTS

Significant differences existed for all the CT quantitative parameters in both groups (p<0.01). The values of area under the curve (AUC) were 0.783 of maximum diameter (95% CI: 0.711-0.843), 0.779 of longest diameter perpendicular to maximum diameter (95% CI: 0.707-0.840), 0.796 of largest cross-sectional area (95% CI: 0.726-0.855), 0.781 of volume (95% CI: 0.710-0.842), 0.794 of mass (95% CI: 0.722-0.865) and 0.625 of mean attenuation value (95% CI: 0.546-0.700), respectively. A pairwise-manner comparison showed the AUC of mean attenuation value was the smallest (p<0.01). Logistic regression analysis showed the largest cross-sectional area (OR=2.307, 95% CI: 1.689-3.150) was the independent predictor for IAC with a cut-off value of 2.22 cm².

CONCLUSIONS

CT quantitative parameters could predict the invasiveness of lung pGGNs. The largest cross-sectional area is the most valuable independent predictor and the mean attenuation value is less valuable.

Radiomic analysis of pulmonary ground-glass opacity nodules for distinction of preinvasive lesions, invasive pulmonary adenocarcinoma and minimally invasive adenocarcinoma based on quantitative texture analysis of CT

Objective

To identify the differences among preinvasive lesions, minimally invasive adenocarcinomas (MIAs) and invasive pulmonary adenocarcinomas (IPAs) based on radiomic feature analysis with computed tomography (CT).

Methods

A total of 109 patients with ground-glass opacity lesions (GGOs) in the lungs determined by CT examinations were enrolled, all of whom had received a pathologic diagnosis. After the manual delineation and segmentation of the GGOs as regions of interest (ROIs), the patients were subdivided into three groups based on pathologic analyses: the preinvasive lesions (including atypical adenomatous hyperplasia and adenocarcinoma in situ) subgroup, the MIA subgroup and the IPA subgroup. Next, we obtained the texture features of the GGOs. The data analysis was aimed at finding both the differences between each pair of the groups and predictors to distinguish any two pathologic subtypes using logistic regression. Finally, a receiver operating characteristic (ROC) curve was applied to accurately evaluate the performances of the regression models.

Results

We found that the voxel count feature (P<0.001) could be used as a predictor for distinguishing IPAs from preinvasive lesions. However, the surface area feature (P=0.040) and the extruded surface area feature (P=0.013) could be predictors of IPAs compared with MIAs. In addition, the correlation feature (P=0.046) could distinguish preinvasive lesions from MIAs better.

Conclusions

Preinvasive lesions, MIAs and IPAs can be discriminated based on texture features within CT images, although the three diseases could all appear as GGOs on CT images. The diagnoses of these three diseases are very important for clinical surgery.