Software performance in segmenting ground-glass and solid components of subsolid nodules in pulmonary adenocarcinomas

Comparative analysis of three-dimensional and two-dimensional models for predicting the malignancy probability of subsolid nodules

AIM

To construct three-dimensional (3D) and two-dimensional (2D) models to predict the malignancy probability of subsolid nodules (SSNs) and compare their effectiveness.

MATERIALS AND METHODS

A total of 371 SSNs from 332 patients, collected between January 2020 and January 2024, were included in the study. The SSNs were divided into a training set for constructing the models and a test set for validating the models. Models were developed using binary logistic backward regression, based on factors that showed significant differences in univariate analyses. The performance of the models was assessed using the area under the curve (AUC) of the receiver operating characteristic (ROC). The AUCs of different models were compared using the DeLong test.

RESULTS

The AUCs for the two 3D models, one 2D model, and the Brock model were 0.785 (0.733-0.836), 0.776 (0.723-0.829), 0.764 (0.710-0.818), and 0.738 (0.679-0.798) in the training set. In the test set, these AUCs were 0.817 (0.706-0.928), 0.796 (0.679-0.913), 0.771 (0.647-0.895), and 0.790 (0.678-0.903). The two 3D models demonstrated statistically significant differences from the Brock model in the training set (P=0.024 and P=0.046). None of the four models showed significant differences in the test set (all P>0.05).

CONCLUSION

The 3D models outperform both the 2D model and the Brock model in predicting the malignancy probability of SSNs, and the 3D model incorporating volume, mean CT attenuation value, and lobulation as factors performed the best.

Subsolid Nodules: Significance and Current Understanding

Subsolid nodules are heterogeneously appearing and behaving entities, commonly encountered incidentally and in high-risk populations. Accurate characterization of subsolid nodules, and application of evolving surveillance guidelines, facilitates evidence-based and multidisciplinary patient-centered management.

Using machine learning-based radiomics to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes

OBJECTIVE

To establish a machine learning-based radiomics model to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes, thereby achieving accurate preoperative classification.

MATERIALS AND METHODS

A retrospective analysis was conducted on MRI T1WI-enhanced images of 105 patients with glioma and 172 patients with solitary brain metastasis from lung cancer, which were confirmed pathologically. The patients were divided into the training group and validation group in an 8:2 ratio for image segmentation, extraction, and filtering; multiple layer perceptron (MLP), support vector machine (SVM), random forest (RF), and logistic regression (LR) were used for modeling; fivefold cross-validation was used to train the model; the validation group was used to evaluate and assess the predictive performance of the model, ROC curve was used to calculate the accuracy, sensitivity, and specificity of the model, and the area under curve (AUC) was used to assess the predictive performance of the model.

RESULTS

The accuracy and AUC of the MLP differentiation model for high-grade glioma and solitary brain metastasis in the validation group was 0.992, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.968, respectively. The accuracy and AUC for the MLP and SVM differentiation model for high-grade glioma and small cell lung cancer brain metastasis in the validation group was 0.966, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.929, respectively. The accuracy and AUC for the MLP differentiation model for high-grade glioma and non-small cell lung cancer brain metastasis in the validation group was 0.982, 0.999, respectively, while the sensitivity and specificity were 0.958, 1.000, respectively.

CONCLUSION

The application of machine learning-based radiomics has a certain clinical value in differentiating glioma from solitary brain metastasis from lung cancer and its subtypes. In the HGG/SBM and HGG/NSCLC SBM validation groups, the MLP model had the best diagnostic performance, while in the HGG/SCLC SBM validation group, the MLP and SVM models had the best diagnostic performance.

An Effective Malignancy Prediction Model for Incidentally Detected Pulmonary Subsolid Nodules Based on Current and Prior CT Scans

INTRODUCTION

It is challenging to diagnose and manage incidentally detected pulmonary subsolid nodules due to their indolent nature and heterogeneity. The objective of this study is to construct a decision tree-based model to predict malignancy of a subsolid nodule based on radiomics features and evolution over time.

MATERIALS AND METHODS

We derived a training set (2947 subsolid nodules), a test set (280 subsolid nodules) from a cohort of outpatient CT scans, and a second test set (5171 subsolid nodules) from the National Lung Cancer Screening Trial (NLST). A Computer-Aided Diagnosis system (CADs) automatically extracted 28 preselected radiomics features, and we calculated the feature change rates as the change of the quantitative measure per time unit between the prior and current CT scans. We built classification models based on XGBoost and employed 5-fold cross validation to optimize the parameters.

RESULTS

The model that combined radiomics features with their change rates performed the best. The Areas Under Curve (AUCs) on the outpatient test set and on the NLST test set were 0.977 (95% CI, 0.958-0.996) and 0.955 (95% CI, 0.930-0.980), respectively. The model performed consistently well on subgroups stratified by nodule diameters, solid components, and CT scan intervals.

CONCLUSION

This decision tree-based model trained with the outpatient dataset gives promising predictive performance on the malignancy of pulmonary subsolid nodules. Additionally, it can assist clinicians to deliver more accurate diagnoses and formulate more in-depth follow-up strategies.

The solid component within part-solid nodules: 3-dimensional quantification, correlation with the malignant grade of nonmucinous pulmonary adenocarcinomas, and comparisons with 2-dimentional measures and semantic features in low-dose computed tomography

BACKGROUND

There is no consensus on 3-dimensional (3D) quantification method for solid component within part-solid nodules (PSNs). This study aimed to find the optimal attenuation threshold for the 3D solid component proportion in low-dose computed tomography (LDCT), namely the consolidation/tumor ratio of volume (CTRV), basing on its correlation with the malignant grade of nonmucinous pulmonary adenocarcinomas (PAs) according to the 5th edition of World Health Organization classification. Then we tested the ability of CTRV to predict high-risk nonmucinous PAs in PSNs, and compare its performance with 2-dimensional (2D) measures and semantic features.

METHODS

A total of 313 consecutive patients with 326 PSNs, who underwent LDCT within one month before surgery and were pathologically diagnosed with nonmucinous PAs, were retrospectively enrolled and were divided into training and testing cohorts according to scanners. The CTRV were automatically generated by setting a series of attenuation thresholds from - 400 to 50 HU with an interval of 50 HU. The Spearman's correlation was used to evaluate the correlation between the malignant grade of nonmucinous PAs and semantic, 2D, and 3D features in the training cohort. The semantic, 2D, and 3D models to predict high-risk nonmucinous PAs were constructed using multivariable logistic regression and validated in the testing cohort. The diagnostic performance of these models was evaluated by the area under curve (AUC) of receiver operating characteristic curve.

RESULTS

The CTRV at attenuation threshold of -250 HU (CTRV_- 250HU) showed the highest correlation coefficient among all attenuation thresholds (r = 0.655, P < 0.001), which was significantly higher than semantic, 2D, and other 3D features (all P < 0.001). The AUCs of CTRV_- 250HU to predict high-risk nonmucinous PAs were 0.890 (0.843-0.927) in the training cohort and 0.832 (0.737-0.904) in the testing cohort, which outperformed 2D and semantic models (all P < 0.05).

CONCLUSIONS

The optimal attenuation threshold was - 250 HU for solid component volumetry in LDCT, and the derived CTRV_- 250HU might be valuable for the risk stratification and management of PSNs in lung cancer screening.

Development and Validation of a Comprehensive Model for Predicting Distant Metastasis of Solid Lung Adenocarcinoma: 3D Radiomics, 2D Radiomics and Clinical Features

OBJECTIVE

To develop and validate models for predicting distant metastases in patients with solid lung adenocarcinomas using 3D radiomic features, 2D radiomic features, clinical features, and their combinations.

METHODS

This retrospective study included 253 eligible patients with solid adenocarcinoma of the lung diagnosed at our hospital between August 2018 and August 2021. 3D and 2D regions of interest were segmented from computed tomography-enhanced thin-slice images of the venous phase, and 851 radiomic features were extracted in each region. The Least Absolute Shrinkage and Selection Operator (LASSO) algorithm was used to select radiomic features and calculate radiomic scores, and logistic regression was used to develop the model. Development of a 3D radiomics model (model 1), a 2D radiomics model (model 2), a combined 3D radiomics and 2D radiomics model (model 3), a clinical model (model 4), and a comprehensive model (model 5) for the prediction of distant metastases in patients with solid lung adenocarcinomas. Nomograms were drawn to illustrate model 5, and receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) were used for model evaluation.

RESULTS

The AUC (area under the curve) of model 1, model 2, model 3, model 4, and model 5 in the test set was 0.711, 0.769, 0.775, 0.829, and 0.892, respectively. The Delong test showed that AUC values were statistically different between model 5 and model 1 (p=0.001), and there was no statistical difference in AUC between the other models. Based on a comprehensive review of DCA, ROC curve, and Akaike information criterion (AIC), Model 5 is demonstrated to have better clinical utility, goodness of fit, and parsimony.

CONCLUSION

A comprehensive model based on 3D radiomic features, 2D radiomic features, and clinical features has the potential to predict distant metastasis in patients with solid lung adenocarcinomas.

Inter-Reader Variability of Volumetric Subsolid Pulmonary Nodule Radiomic Features

OBJECTIVE

To evaluate the inter-observer consistency for subsolid pulmonary nodule radiomic features.

MATERIALS AND METHODS

Subsolid nodules were selected by reviewing radiology reports of CT examinations performed December 1, 2015 to April 1, 2016. Patients with CTs at two time points were included in this study. There were 55 patients with subsolid nodules, of whom 14 had two nodules. Of 69 subsolid nodules, 66 were persistent at the second time point, yielding 135 lesions for segmentation. Two thoracic radiologists and an imaging fellow segmented the lesions using a semi-automated volumetry algorithm (Syngo.via Vb20, Siemens). Coefficient of variation (CV) was used to assess consistency of 91 quantitative measures extracted from the subsolid nodule segmentations, including first and higher order texture features. The accuracy of segmentation was visually graded by an experienced thoracic radiologist. Influencing factors on radiomic feature consistency and segmentation accuracy were assessed using generalized estimating equation analyses and the Exact Mann-Whitney test.

RESULTS

Mean patient age was 71 (38-93 years), with 39 women and 16 men. Mean nodule volume was 1.39mL, range .03-48.2mL, for 135 nodules. Several radiomic features showed high inter-reader consistency (CV<5%), including entropy, uniformity, sphericity, and spherical disproportion. Descriptors such as surface area and energy had low consistency across inter-reader segmentations (CV>10%). Nodule percent solid component and attenuation influenced inter-reader variability of some radiomic features. The presence of contrast did not significantly affect the consistency of subsolid nodule radiomic features. Near perfect segmentation, within 5% of actual nodule size, was achieved in 68% of segmentations, and very good segmentation, within 25% of actual nodule size, in 94%. Morphologic features including nodule margin and shape (each p <0.01), and presence of air bronchograms (p = 0.004), bubble lucencies (p = 0.02) and broad pleural contact (p < 0.01) significantly affected the probability of near perfect segmentation. Stroke angle (p = 0.001) and length (p < 0.001) also significantly influenced probability of near perfect segmentation.

CONCLUSIONS

The inter-observer consistency of radiomic features for subsolid pulmonary nodules varies, with high consistency for several features, including sphericity, spherical disproportion, and first and higher order entropy, and normalized non-uniformity. Nodule morphology influences the consistency of subsolid nodule radiomic features, and the accuracy of subsolid nodule segmentation.

Quantitative CT for detecting COVID‑19 pneumonia in suspected cases

Background

Corona Virus Disease 2019 (COVID-19) is currently a worldwide pandemic and has a huge impact on public health and socio-economic development. The purpose of this study is to explore the diagnostic value of the quantitative computed tomography (CT) method by using different threshold segmentation techniques to distinguish between patients with or without COVID-19 pneumonia.

Methods

A total of 47 patients with suspected COVID-19 were retrospectively analyzed, including nine patients with positive real-time fluorescence reverse transcription polymerase chain reaction (RT-PCR) test (confirmed case group) and 38 patients with negative RT-PCR test (excluded case group). An improved 3D convolutional neural network (VB-Net) was used to automatically extract lung lesions. Eight different threshold segmentation methods were used to define the ground glass opacity (GGO) and consolidation. The receiver operating characteristic (ROC) curves were used to compare the performance of various parameters with different thresholds for diagnosing COVID-19 pneumonia.

Results

The volume of GGO (VOGGO) and GGO percentage in the whole lung (GGOPITWL) were the most effective values for diagnosing COVID-19 at a threshold of − 300 HU, with areas under the curve (AUCs) of 0.769 and 0.769, sensitivity of 66.67 and 66.67%, specificity of 94.74 and 86.84%. Compared with VOGGO or GGOPITWL at a threshold of − 300 Hounsfield units (HU), the consolidation percentage in the whole lung (CPITWL) with thresholds at − 400 HU, − 350 HU, and − 250 HU were statistically different. There were statistical differences in the infection volume and percentage of the whole lung, right lung, and lobes between the two groups. VOGGO, GGOPITWL, and volume of consolidation (VOC) were also statistically different at the threshold of − 300 HU.

Conclusions

Quantitative CT provides an image quantification method for the auxiliary diagnosis of COVID-19 and is expected to assist in confirming patients with COVID-19 pneumonia in suspected cases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-021-06556-z.

Semi-Quantitative Analysis for Determining the Optimal Threshold Value on CT to Measure the Solid Portion of Pulmonary Subsolid Nodules

PURPOSE

This study aimed to investigate the optimal threshold value in Hounsfield units (HU) on CT to detect the solid components of pulmonary subsolid nodules using pathologic invasive foci as reference.

MATERIALS AND METHODS

Thin-section non-enhanced chest CT scans of 25 patients with pathologically confirmed minimally invasive adenocarcinoma were retrospectively reviewed. On CT images, the solid portion was defined as the area with higher attenuation than various HU thresholds ranging from -600 to -100 HU in 50-HU intervals. The solid portion was measured as the largest diameter on axial images and as the maximum diameter on multiplanar reconstruction images. A linear mixed model was used to evaluate bias in each threshold by using the pathological size of invasive foci as reference.

RESULTS

At a threshold of -400 HU, the biases were lowest between the largest/maximum diameter of the solid portion of subsolid nodule and the size of invasive foci of the pathological specimen, with 0.388 and -0.0176, respectively. They showed insignificant difference (p = 0.2682, p = 0.963, respectively) at a threshold of -400 HU.

CONCLUSION

For quantitative analysis, -400 HU may be the optimal threshold to define the solid portion of subsolid nodules as a surrogate marker of invasive foci.

Generalized chest CT and lab curves throughout the course of COVID-19

A better understanding of temporal relationships between chest CT and labs may provide a reference for disease severity over the disease course. Generalized curves of lung opacity volume and density over time can be used as standardized references from well before symptoms develop to over a month after recovery, when residual lung opacities remain. 739 patients with COVID-19 underwent CT and RT-PCR in an outbreak setting between January 21st and April 12th, 2020. 29 of 739 patients had serial exams (121 CTs and 279 laboratory measurements) over 50 ± 16 days, with an average of 4.2 sequential CTs each. Sequential volumes of total lung, overall opacity and opacity subtypes (ground glass opacity [GGO] and consolidation) were extracted using deep learning and manual segmentation. Generalized temporal curves of CT and laboratory measurements were correlated. Lung opacities appeared 3.4 ± 2.2 days prior to symptom onset. Opacity peaked 1 day after symptom onset. GGO onset was earlier and resolved later than consolidation. Lactate dehydrogenase, and C-reactive protein peaked earlier than procalcitonin and leukopenia. The temporal relationships of quantitative CT features and clinical labs have distinctive patterns and peaks in relation to symptom onset, which may inform early clinical course in patients with mild COVID-19 pneumonia, or may shed light upon chronic lung effects or mechanisms of medical countermeasures in clinical trials.

Use of a Commercially Available Deep Learning Algorithm to Measure the Solid Portions of Lung Cancer Manifesting as Subsolid Lesions at CT: Comparisons with Radiologists and Invasive Component Size at Pathologic Examination

Background The solid portion size of lung cancer lesions manifesting as subsolid lesions is key in their management, but the automatic measurement of such lesions by means of a deep learning (DL) algorithm needs evaluation. Purpose To evaluate the performance of a commercially available DL algorithm for automatic measurement of the solid portion of surgically proven lung adenocarcinomas manifesting as subsolid lesions. Materials and Methods Surgically proven lung adenocarcinomas manifesting as subsolid lesions on CT images between January 2018 and December 2018 were retrospectively included. Five radiologists independently measured the maximal axial diameter of the solid portion of lesions. The DL algorithm automatically segmented and measured the maximal axial diameter of the solid portion. Reader measurements, software measurements, and invasive component size at pathologic examination were compared by using intraclass correlation coefficient (ICC) and Bland-Altman plots. Results A total of 448 patients (mean age, 63 years ± 10 [standard deviation]; 264 women) with 448 lesions were evaluated (invasive component size, 3-65 mm). The measurement agreements between each radiologist and the DL algorithm were very good (ICC range, 0.82-0.89). When a radiologist was replaced with the DL algorithm, the ICCs ranged from 0.87 to 0.90, with an ICC of 0.90 among five radiologists. The mean difference between the DL algorithm and each radiologist ranged from -3.7 to 1.5 mm. The widest 95% limit of agreement between the DL algorithm and each radiologist (-15.7 to 8.3 mm) was wider than pairwise comparisons of radiologists (-7.7 to 13.0 mm). The agreement between the DL algorithm and invasive component size at pathologic evaluation was good, with an ICC of 0.67. Measurements by the DL algorithm (mean difference, -6.0 mm) and radiologists (mean difference, -7.5 to -2.3 mm) both underestimated invasive component size. Conclusion Automatic measurements of solid portions of lung cancer manifesting as subsolid lesions by the deep learning algorithm were comparable with manual measurements and showed good agreement with invasive component size at pathologic evaluation. © RSNA, 2021 Online supplemental material is available for this article.

Severity assessment of COVID-19 using CT image features and laboratory indices

The coronavirus disease 2019 (COVID-19) is now a global pandemic. Tens of millions of people have been confirmed with infection, and also more people are suspected. Chest computed tomography (CT) is recognized as an important tool for COVID-19 severity assessment. As the number of chest CT images increases rapidly, manual severity assessment becomes a labor-intensive task, delaying appropriate isolation and treatment. In this paper, a study of automatic severity assessment for COVID-19 is presented. Specifically, chest CT images of 118 patients (age 46.5 ± 16.5 years, 64 male and 54 female) with confirmed COVID-19 infection are used, from which 63 quantitative features and 110 radiomics features are derived. Besides the chest CT image features, 36 laboratory indices of each patient are also used, which can provide complementary information from a different view. A random forest (RF) model is trained to assess the severity (non-severe or severe) according to the chest CT image features and laboratory indices. Importance of each chest CT image feature and laboratory index, which reflects the correlation to the severity of COVID-19, is also calculated from the RF model. Using three-fold cross-validation, the RF model shows promising results: 0.910 (true positive ratio), 0.858 (true negative ratio) and 0.890 (accuracy), along with AUC of 0.98. Moreover, several chest CT image features and laboratory indices are found to be highly related to COVID-19 severity, which could be valuable for the clinical diagnosis of COVID-19.

A novel deep learning-based quantification of serial chest computed tomography in Coronavirus Disease 2019 (COVID-19)

This study aims to explore and compare a novel deep learning-based quantification with the conventional semi-quantitative computed tomography (CT) scoring for the serial chest CT scans of COVID-19. 95 patients with confirmed COVID-19 and a total of 465 serial chest CT scans were involved, including 61 moderate patients (moderate group, 319 chest CT scans) and 34 severe patients (severe group, 146 chest CT scans). Conventional CT scoring and deep learning-based quantification were performed for all chest CT scans for two study goals: (1) Correlation between these two estimations; (2) Exploring the dynamic patterns using these two estimations between moderate and severe groups. The Spearman's correlation coefficient between these two estimation methods was 0.920 (p < 0.001). predicted pulmonary involvement (CT score and percent of pulmonary lesions calculated using deep learning-based quantification) increased more rapidly and reached a higher peak on 23rd days from symptom onset in severe group, which reached a peak on 18th days in moderate group with faster absorption of the lesions. The deep learning-based quantification for COVID-19 showed a good correlation with the conventional CT scoring and demonstrated a potential benefit in the estimation of disease severities of COVID-19.

Shades of Gray: Subsolid Nodule Considerations and Management

Subsolid nodules are common on chest CT imaging and may be either benign or malignant. Their varied features and broad differential diagnoses present management challenges. Although subsolid nodules often represent lung adenocarcinomas, other possibilities are common and influence management. Practice guidelines exist for subsolid nodule management for both incidentally and screening-detected nodules, incorporating patient and nodule characteristics. This review highlights the similarities and differences among these algorithms, with the intent of providing a resource for comparison and aid in choosing management options.

A predictive model for disease progression in non-severely ill patients with coronavirus disease 2019

Over the past 3 months, coronavirus disease 2019 (COVID-19) has emerged across China and developed into a worldwide outbreak [1]. The disease has caused varying degrees of illness. The proportion of patients with COVID-19 with non-severe illness was 84.3% on admission, and severe cases accounted for 15.7% [2]. Most of the non-severe pneumonia patients would gradually alleviate and be cured with treatment, while others would rapidly progress to severe illness, which has a poor prognosis [3, 4]. As recently reported, the cumulative risk of the composite end-point was 3.6% in all COVID-19 patients, and the cumulative risk was 20.6% for severe illness [2].

A predictive model for COVID-19 https://bit.ly/2WHiEkJ

Radiological-pathological correlation of subsolid pulmonary nodules: A single centre retrospective evaluation of the 2011 IASLC adenocarcinoma classification system

INTRODUCTION

The 2011 IASLC classification system proposes guidelines for radiologists and pathologists to classify adenocarcinomas spectrum lesions as preinvasive, minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IA). IA portends the worst clinical prognosis, and the imaging distinction between MIA and IA is controversial.

MATERIALS AND METHODS

Subsolid pulmonary nodules resected by microcoil localization over a three-year period were retrospectively reviewed by three chest radiologists and a pulmonary pathologist. Nodules were classified radiologically based on preoperative computed tomography (CT), with the solid nodule component measured on mediastinal windows applied to high-frequency lung kernel reconstructions, and pathologically according to 2011 IASLC criteria. Radiology interobserver and radiological-pathological variability of nodule classification, and potential reasons for nodule classification discordance were assessed.

RESULTS

Seventy-one subsolid nodules in 67 patients were included. The average size of invasive disease focus at histopathology was 5 mm (standard deviation 5 mm). Radiology interobserver agreement of nodule classification was good (Cohen's Kappa = 0.604, 95 % CI: 0.447 to 0.761). Agreement between consensus radiological interpretation and pathological category was fair (Cohen's Kappa = 0.236, 95 % CI: 0.054-0.421). Radiological and pathological nodule classification were concordant in 52 % (37 of 71) of nodules. The IASLC proposed CT solid component cut-off of 5 mm to distinguish MIA and IA yielded a sensitivity of 59 % and specificity of 80 %. Common reasons for nodule classification discordance included multiple solid components within a nodule on CT, scar and stromal collapse at pathology, and measurement variability.

CONCLUSION

Solid component(s) within persistent part-solid pulmonary nodules raise suspicion for invasive adenocarcinoma. Preoperative imaging classification is frequently discordant from final pathology, reflecting interpretive and technical challenges in radiological and pathological analysis.

Characteristics of disease progress in patients with coronavirus disease 2019 in Wuhan, China

Coronavirus disease 2019 (COVID-19) patients were classified into four clinical stages (uncomplicated illness, mild, severe and critical pneumonia) depending on disease severity. We aim to investigate the corresponding clinical, radiological and laboratory characteristics between different clinical stages. A retrospective, single-centre study of 101 confirmed patients with COVID-19 at Renmin Hospital of Wuhan University from 2 January to 28 January 2020 was enrolled; follow-up endpoint was on 8 February 2020. Clinical data were collected and compared during the course of illness. The median age of the 101 patients was 51.0 years and 33.6% were medical staff. Fever (68%), cough (50%) and fatigue (23%) are the most common symptoms. About 26% patients underwent the mechanical ventilation and 98% patients were treated with antibiotics. Thirty-seven per cent patients were cured and 11 died. On admission, the number of patients with uncomplicated illness, mild, severe and critical pneumonia were 2 [2%], 86 [85%], 11 [11%] and 2 [2%]. Forty-four of the 86 mild pneumonia progressed to severe illness within 4 days, with nine patients worsened due to critical pneumonia within 4 days. Two of the 11 severe patients improved to mild condition while three others deteriorated. Significant differences were observed among groups of different clinical stages in numbers of influenced pulmonary segments (6 vs. 12 vs. 17, P < 0.001). A significantly upward trend was witnessed in ground-glass opacities overlapped with striped shadows (33% vs. 42% vs. 55% vs. 80%, P < 0.001), while pure ground-glass opacities gradually decreased as disease progressed (45% vs. 35% vs. 24% vs. 13%, P < 0.001) within 12 days. Lymphocytes, prealbumin and albumin showed a downtrend as disease progressed from mild to severe or critical condition, an uptrend was found in white blood cells, C-reactive protein, neutrophils and lactate dehydrogenase. The proportions of serum amyloid A > 300 mg/l in mild, severe and critical conditions were 18%, 46% and 71%, respectively.

Optimization of CT windowing for diagnosing invasiveness of adenocarcinoma presenting as sub-solid nodules

PURPOSE

To evaluate the optimal window setting to diagnose the invasiveness of lung adenocarcinoma in sub-solid nodules (SSNs).

METHODS

We retrospectively included 437 SSNs and randomly divided them 3:1 into a training group (327) and a testing group (110). The presence of a solid component was regarded as indicator of invasiveness. At fixed window level (WL) of 35 Hounsfield Units (HU), two readers adjusted the window width (WW) in the training group and recorded once a solid component appeared or disappeared on CT images acquired at 120 kVp. The optimal WW cut-off value to differentiate between invasive and pre-invasive lesions, based on the receiver operating characteristic (ROC) curve, was defined as "core" WW. The diagnostic performances of the mediastinal window setting (WW/WL, 350/35 HU) and core window setting were then compared in the testing group.

RESULTS

Of the 437 SSNs, 88 were pre-invasive [17 atypical adenomatous hyperplasia (AAH) and 71 adenocarcinoma in situ (AIS)], 349 were invasive [233 minimally invasive adenocarcinoma (MIA), 116 invasive adenocarcinoma (IA)]. In training group, the core WW of 1175 HU was the optimal cut-off to detect solid components of SSNs (AUC:0.79). In testing group, the sensitivity, specificity, positive, negative predictive value, and diagnostic accuracy for SSN invasiveness were 49.4%, 90.5%, 95.7%, 29.7%, and 57.3% for mediastinal window setting, and 87.6%, 76.2%, 91.6%, 76.2%, and 85.5% for core window setting.

CONCLUSION

At 120 kVp, core window setting (WW/WL, 1175/35 HU) outperformed the traditional mediastinal window setting to diagnose the invasiveness of SSNs.

A deep residual learning network for predicting lung adenocarcinoma manifesting as ground-glass nodule on CT images

OBJECTIVE

To develop a deep learning-based artificial intelligence (AI) scheme for predicting the likelihood of the ground-glass nodule (GGN) detected on CT images being invasive adenocarcinoma (IA) and also compare the accuracy of this AI scheme with that of two radiologists.

METHODS

First, we retrospectively collected 828 histopathologically confirmed GGNs of 644 patients from two centers. Among them, 209 GGNs are confirmed IA and 619 are non-IA, including 409 adenocarcinomas in situ and 210 minimally invasive adenocarcinomas. Second, we applied a series of pre-preprocessing techniques, such as image resampling, rescaling and cropping, and data augmentation, to process original CT images and generate new training and testing images. Third, we built an AI scheme based on a deep convolutional neural network by using a residual learning architecture and batch normalization technique. Finally, we conducted an observer study and compared the prediction performance of the AI scheme with that of two radiologists using an independent dataset with 102 GGNs.

RESULTS

The new AI scheme yielded an area under the receiver operating characteristic curve (AUC) of 0.92 ± 0.03 in classifying between IA and non-IA GGNs, which is equivalent to the senior radiologist's performance (AUC 0.92 ± 0.03) and higher than the score of the junior radiologist (AUC 0.90 ± 0.03). The Kappa value of two sets of subjective prediction scores generated by two radiologists is 0.6.

CONCLUSIONS

The study result demonstrates using an AI scheme to improve the performance in predicting IA, which can help improve the development of a more effective personalized cancer treatment paradigm.

KEY POINTS

• The feasibility of using a deep learning method to predict the likelihood of the ground-glass nodule being invasive adenocarcinoma. • Residual learning-based CNN model improves the performance in classifying between IA and non-IA nodules. • Artificial intelligence (AI) scheme yields higher performance than radiologists in predicting invasive adenocarcinoma.

Validation of lesion simulations in clinical CT data for anonymized chest and abdominal CT databases

PURPOSE

To make available to the medical imaging community a computed tomography (CT) image database composed of hybrid datasets (patient CT images with digitally inserted anthropomorphic lesions) where lesion ground truth is known a priori. It is envisioned that such a dataset could be a resource for the assessment of CT image quality, machine learning, and imaging technologies [e.g., computer aided detection (CAD) and segmentation algorithms].

ACQUISITION AND VALIDATION METHODS

This HIPPA compliant, IRB waiver of approval study consisted of utilizing 120 chest and 100 abdominal clinically acquired adult CT exams. One image series per patient exam was utilized based on coverage of the anatomical region of interest (either the thorax or abdomen). All image series were de-identified. Simulated lesions were derived from a library of anatomically informed digital lesions (93 lung and 50 liver lesions) where six and four digital lesions with nominal diameters ranging from 4 to 20 mm were inserted into lung and liver image series, respectively. Locations for lesion insertion were randomly chosen. A previously validated lesion simulation and virtual insertion technique were utilized. The resulting hybrid images were reviewed by three experienced radiologists to assure similarity with routine clinical imaging in a diverse adult population.

DATA FORMAT AND USAGE NOTES

The database is composed of four datasets that contain 100 patient cases each, for a total of 400 image series accompanied by Matlab.mat tables that provide descriptive information about the virtually inserted lesions (i.e., size, shape, opacity, and insertion location in physical (world) coordinates and voxel indices). All image and metadata are stored in DICOM format on the Quantitative Imaging Data Warehouse (https://qidw.rsna.org/#collection/57d463471cac0a4ec8ff8f46/folder/5b23dceb1cac0a4ec800a770?dialog=login), in two sets: (a) QIBA CT Hybrid Dataset I which contains Lung I and Liver I datasets, and (b) QIBA CT Hybrid Dataset II which contains Lung II and Liver II datasets. The QIDW is supported by the Radiological Society of North America (RSNA). Registration is required upon initial log in.

POTENTIAL APPLICATIONS

By simulating lesion opacity (full solid, part solid and ground glass), size, and texture, the relationship between lesion morphology and segmentation or CAD algorithm performance can be investigated without the need for repetitive patient exams. This database can also serve as a reference standard for device and reader performance studies.

High versus low attenuation thresholds to determine the solid component of ground-glass opacity nodules

Objectives

To evaluate and compare the diagnostic accuracy of high versus low attenuation thresholds for determining the solid component of ground-glass opacity nodules (GGNs) for the differential diagnosis of adenocarcinoma in situ (AIS) from minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IA).

Methods

Eighty-six pathologically confirmed GGNs < 3 cm observed in 86 patients (27 male, 59 female; mean age, 59.3 ± 11.0 years) between January 2013 and December 2015 were retrospectively included. The solid component of each GGN was defined using two different attenuation thresholds: high (-160 Hounsfield units [HU]) and low (-400 HU). According to the presence or absence of solid portions, each GGN was categorized as a pure GGN or part-solid GGN. Solid components were regarded as indicators of invasive foci, suggesting MIA or IA.

Results

Among the 86 GGNs, there were 57 cases of IA, 19 of MIA, and 10 of AIS. Using the high attenuation threshold, 44 were categorized as pure GGNs and 42 as part-solid GGNs. Using the low attenuation threshold, 13 were categorized as pure GGNs and 73 as part-solid GGNs. The sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy for the invasive focus were 55.2%, 100%, 100%, 22.7%, and 60.4%, respectively, for the high attenuation threshold, and 93.4%, 80%, 97.2%, 61.5%, and 91.8%, respectively, for the low attenuation threshold.

Conclusion

The low attenuation threshold was better than the conventional high attenuation threshold for determining the solid components of GGNs, which indicate invasive foci.

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.

A simple prediction model using size measures for discrimination of invasive adenocarcinomas among incidental pulmonary subsolid nodules considered for resection

OBJECTIVES

To develop and validate a concise prediction model using simple size measures for the discrimination of invasive pulmonary adenocarcinomas (IPAs) among incidentally detected subsolid nodules (SSNs) considered for resection and to compare its diagnostic performance with the Brock model.

METHODS

This retrospective institutional review board-approved study included 427 surgically resected SSNs (121 preinvasive lesions/minimally invasive adenocarcinomas [MIAs] and 306 IPAs) from 407 patients. After stratified random splitting of the study population into the training and validation sets (3:1), a simple logistic model was constructed using nodule size, solid proportion, and type for the differentiation of IPAs. Diagnostic performance of this model was compared with the original and modified Brock models using the DeLong method for area under the receiver-operating characteristic curve (AUC) and McNemar test for diagnostic sensitivity and specificity.

RESULTS

Our proposed model had an AUC of 0.859 in the validation set, while the original Brock model showed an AUC of 0.775 (p = 0.035) and the modified Brock model exhibited an AUC of 0.787 (p = 0.006). At equally high specificity of 90%, our proposed model exhibited significantly higher sensitivity (65.8%) than the original and modified Brock models (38.2% and 50.0%; p < 0.001 and 0.008, respectively).

CONCLUSIONS

Our study results demonstrated that the proposed concise model outperformed both Brock models, demonstrating its potential to be utilized as a specific tool to differentiate IPAs from preinvasive lesions and MIAs, which were considered for resection. External validation studies are warranted for the population with incidentally detected SSNs including small SSNs to confirm our observations.

KEY POINTS

• Size measures provided sufficient information for the risk stratification of surgical candidate incidental subsolid nodules. • Our proposed concise model showed higher diagnostic performance than the Brock model for incidentally detected subsolid nodules. • Our proposed model can specifically differentiate invasive adenocarcinomas among incidentally detected subsolid nodules and reduce overtreatment for indolent subsolid nodules.

Implication of total tumor size on the prognosis of patients with clinical stage IA lung adenocarcinomas appearing as part-solid nodules: Does only the solid portion size matter?

OBJECTIVES

The aim was to investigate the effect of clinico-radiologic variables, including total tumor (T_total) size and clinical T category, on the prognosis of patients with stage IA (T1N0M0) lung adenocarcinomas appearing as part-solid nodules (PSNs).

METHODS

This institutional review board-approved retrospective study included 506 patients (male:female = 200:306; median age, 62 years) with PSNs of the adenocarcinoma spectrum in clinical stage IA who underwent standard lobectomy at a single tertiary medical center. Prognostic stratification of the patients in terms of disease-free survival was analyzed with variables including age, sex, T_total size, solid portion size, clinical T category, and tumor location using univariate and subsequent multivariate Cox regression analysis. Subgroup analysis was performed to reveal the effect of the T_total size at each clinical T category.

RESULTS

Multivariate Cox regression analysis demonstrated that T_total size*cT1b [interaction term; hazard ratio (HR) = 1.091; 95% confidence interval (CI): 1.015, 1.173; p = 0.019] and cT1c (HR = 68.436; 95% CI: 2.797, 1674.415; p = 0.010) were independent risk factors for the tumor recurrence. When patients with cT1b were dichotomized based on a T_total size cutoff of 3.0 cm, PSNs with T_total > 3.0 cm showed a significantly worse outcome (HR = 3.796; 95% CI: 1.006, 14.317; p = 0.049). No significant difference was observed in the probability of recurrence between cT1b with T_total > 3.0 cm and cT1c (p = 0.915).

CONCLUSIONS

T_total size is a significant prognostic factor in adenocarcinoma patients in cT1b without lymph node or distant metastasis. PSNs in cT1b with T_total > 3.0 cm have a comparable risk of lung cancer recurrence to those in cT1c.

KEY POINTS

• Current T descriptor was a powerful prognostic factor in stage IA adenocarcinomas appearing as part-solid nodules. • Total tumor size further stratified risk of recurrence of adenocarcinomas in cT1b. • Upstaging of tumors in cT1b with total tumor size > 3.0 cm may be more appropriate.

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.

Pulmonary subsolid nodules: value of semi-automatic measurement in diagnostic accuracy, diagnostic reproducibility and nodule classification agreement

OBJECTIVES

We hypothesized that semi-automatic diameter measurements would improve the accuracy and reproducibility in discriminating preinvasive lesions and minimally invasive adenocarcinomas from invasive pulmonary adenocarcinomas appearing as subsolid nodules (SSNs) and increase the reproducibility in classifying SSNs.

METHODS

Two readers independently performed semi-automatic and manual measurements of the diameters of 102 SSNs and their solid portions. Diagnostic performance in predicting invasive adenocarcinoma based on diameters was tested using logistic regression analysis with subsequent receiver operating characteristic curves. Inter- and intrareader reproducibilities of diagnosis and SSN classification according to Fleischner's guidelines were investigated for each measurement method using Cohen's κ statistics.

RESULTS

Semi-automatic effective diameter measurements were superior to manual average diameters for the diagnosis of invasive adenocarcinoma (AUC, 0.905-0.923 for semi-automatic measurement and 0.833-0.864 for manual measurement; p<0.05). Reproducibility of diagnosis between the readers also improved with semi-automatic measurement (κ=0.924 for semi-automatic measurement and 0.690 for manual measurement, p=0.012). Inter-reader SSN classification reproducibility was significantly higher with semi-automatic measurement (κ=0.861 for semi-automatic measurement and 0.683 for manual measurement, p=0.022).

CONCLUSIONS

Semi-automatic effective diameter measurement offers an opportunity to improve diagnostic accuracy and reproducibility as well as the classification reproducibility of SSNs.

KEY POINTS

• Semi-automatic effective diameter measurement improves the diagnostic accuracy for pulmonary subsolid nodules. • Semi-automatic measurement increases the inter-reader agreement on the diagnosis for subsolid nodules. • Semi-automatic measurement augments the inter-reader reproducibility for the classification of subsolid nodules.

Quality assurance and quantitative imaging biomarkers in low-dose CT lung cancer screening

After years of assessment through controlled clinical trials, low-dose CT screening for lung cancer is becoming part of clinical practice. As with any cancer screening test, those undergoing lung cancer screening are not being evaluated for concerning signs or symptoms, but are generally in good health and proactively trying to prevent premature death. Given the resultant obligation to achieve the screening aim of early diagnosis while also minimizing the potential for morbidity from workup of indeterminate but ultimately benign screening abnormalities, careful implementation of screening with conformance to currently recognized best practices and a focus on quality assurance is essential. In this review, we address the importance of each component of the screening process to optimize the effectiveness of CT screening, discussing options for quality assurance at each step. We also discuss the potential added advantages, quality assurance requirements and current status of quantitative imaging biomarkers related to lung cancer screening. Finally, we highlight suggestions for improvements and needs for further evidence in evaluating the performance of CT screening as it transitions from the research trial setting into daily clinical practice.

[Threshold Segmentation of Pulmonary Subsolid Nodules on CT Images: Detection and Quantification of the Solid Component]

BACKGROUND

The detection and quantification of solid components in pulmonary subsolid nodules (SSN) are of vital importance on differential diagnosis, pathological speculation and prognosis prediction. However, no objective and wide-accepted criterion has been built up to now. The purpose of this study is to explore the optimal threshold that can be used for the detection and quantification of solid components in SSNs by using threshold segmentation method on computed tomography (CT) images.

METHODS

CT images of 102 SSNs were retrospectively analyzed. To establish a reference standard, the observers made judgments on whether the solid component existed in every SSN and did manual measurements of the volume of solid component with the help of software. Threshold segmentations of every nodule were then performed using different threshold settings and all of the measured volumes were assumed to be solid volumes, then solid-to-total volume ratios were calculated. The results were compared with the reference standards using the receiver operating characteristic curve and Wilcoxon test.

RESULTS

The application of thresholds as -250 HU or -300 HU resulted in high diagnostic value on the detection of solid component, with area under curve values as 0.982 and 0.977, respectively; the cut-off values of solid-to-total volume ratio were 1.10% and 6.14%, respectively; the median volumes of solid components were 202.7 mm3 (598.2 mm3), 247.1 mm3(696.0 mm3), which were not significantly different from the reference standard[199.5 mm3 (743.1 mm3)](P=0.125,1, 0.061,3).

CONCLUSIONS

Threshold segmentation on chest CT images is valuable to detect and quantify the solid component on SSNs, the thresholds as -250 HU and -300 HU are recommended.
.

Lung adenocarcinomas: correlation of computed tomography and pathology findings

Adenocarcinoma is the most common histologic type of lung cancer. Recent lung adenocarcinoma classifications from the International Association for the Study of Lung cancer, the American Thoracic Society and the European Respiratory Society (IASLC/ETS/ERS, 2011) and World Health Organization (WHO, 2015) define a wide range of adenocarcinoma types and subtypes featuring different prognosis and management. This spectrum of lesions translates into various CT presentations and features, which generally show good correlation with histopathology, stressing the key role of the radiologist in the diagnosis and management of those patients. This review aims at helping radiologists to understand the basics of the up-to-date adenocarcinoma pathological classifications, radio-pathological correlations and how to use them in the clinical setting, as well as other imaging-related correlations (radiogenomics, quantitative analysis, PET-CT).