CLSSL-ResNet: Predicting malignancy of solitary pulmonary nodules from CT images by chimeric label with self-supervised learning.

BACKGROUND: Pulmonary granulomatous nodules (GN) with spiculation or lobulation have a similar morphological appearance to solid lung adenocarcinoma (SADC) under computed tomography (CT). However, these two kinds of solid pulmonary nodules (SPN) have different malignancies and are sometimes misdiagnosed. OBJECTIVE: This study aims to predict malignancies of SPNs by a deep learning model automatically. METHODS: A chimeric label with self-supervised learning (CLSSL) is proposed to pre-train a ResNet-based network (CLSSL-ResNet) for distinguishing isolated atypical GN from SADC in CT images. The malignancy, rotation, and morphology labels are integrated into a chimeric label and utilized to pre-train a ResNet50. The pre-trained ResNet50 is then transferred and fine-tuned to predict the malignancy of SPN. Two image datasets of 428 subjects (Dataset1, 307; Dataset2, 121) from different hospitals are collected. Dataset1 is divided into training, validation, and test data by a ratio of 7:1:2 to develop the model. Dataset2 is utilized as an external validation dataset. RESULTS: CLSSL-ResNet achieves an area under the ROC curve (AUC) of 0.944 and an accuracy (ACC) of 91.3%, which was much higher than that of the consensus of two experienced chest radiologists (77.3%). CLSSL-ResNet also outperforms other self-supervised learning models and many counterparts of other backbone networks. In Dataset2, AUC and ACC of CLSSL-ResNet are 0.923 and 89.3%, respectively. Additionally, the ablation experiment result indicates higher efficiency of the chimeric label. CONCLUSION: CLSSL with morphology labels can increase the ability of feature representation by deep networks. As a non-invasive method, CLSSL-ResNet can distinguish GN from SADC via CT images and may support clinical diagnoses after further validation.

Deep radiomic model based on the sphere-shell partition for predicting treatment response to chemotherapy in lung cancer.

BACKGROUND: The prognosis of chemotherapy is important in clinical decision-making for non-small cell lung cancer (NSCLC) patients. OBJECTIVES: To develop a model for predicting treatment response to chemotherapy in NSCLC patients from pre-chemotherapy CT images. MATERIALS AND METHODS: This retrospective multicenter study enrolled 485 patients with NSCLC who received chemotherapy alone as a first-line treatment. Two integrated models were developed using radiomic and deep-learning-based features. First, we partitioned pre-chemotherapy CT images into spheres and shells with different radii around the tumor (0-3, 3-6, 6-9, 9-12, 12-15 mm) containing intratumoral and peritumoral regions. Second, we extracted radiomic and deep-learning-based features from each partition. Third, using radiomic features, five sphere-shell models, one feature fusion model, and one image fusion model were developed. Finally, the model with the best performance was validated in two cohorts. RESULTS: Among the five partitions, the model of 9-12 mm achieved the highest area under the curve (AUC) of 0.87 (95% confidence interval: 0.77-0.94). The AUC was 0.94 (0.85-0.98) for the feature fusion model and 0.91 (0.82-0.97) for the image fusion model. For the model integrating radiomic and deep-learning-based features, the AUC was 0.96 (0.88-0.99) for the feature fusion method and 0.94 (0.85-0.98) for the image fusion method. The best-performing model had an AUC of 0.91 (0.81-0.97) and 0.89 (0.79-0.93) in two validation sets, respectively. CONCLUSIONS: This integrated model can predict the response to chemotherapy in NSCLC patients and assist physicians in clinical decision-making.

The Computed Tomography Findings and Follow-up Course of Pulmonary Nocardiosis.

OBJECTIVE: Our study aimed to elucidate the computed tomography (CT) features and follow-up course of pulmonary nocardiosis patients to improve the understanding and diagnostic accuracy of this disease. METHODS: The chest CT findings and clinical data of patients diagnosed with pulmonary nocardiosis by culture or histopathological examination in our hospital between 2010 and 2019 were retrospectively analyzed. RESULTS: A total of 34 cases of pulmonary nocardiosis were included in our study. Thirteen patients were on long-term immunosuppressant therapy, among whom 6 had disseminated nocardiosis. Among the immunocompetent patients, 16 had chronic lung diseases or a history of trauma. Multiple or solitary nodules represented the most common CT feature (n = 32, 94.12%), followed by ground-glass opacities (n = 26, 76.47%), patchy consolidations (n = 25, 73.53%), cavitations (n = 18, 52.94%), and masses (n = 11, 32.35%). There were 20 cases (61.76%) with mediastinal and hilar lymphadenopathy, 18 (52.94%) with pleural thickening, 15 (44.12%) with bronchiectasis, and 13 (38.24%) with pleural effusion. Significantly higher rates of cavitations were observed among immunosuppressed patients (85% vs 29%, P = 0.005). At follow-up, 28 patients (82.35%) clinically improved with treatment, while 5 (14.71%) had disease progression, and 1 (2.94%) died. CONCLUSIONS: Chronic structural lung diseases and long-term immunosuppressant use were found as risk factors for pulmonary nocardiosis. While the CT manifestations were highly heterogeneous, clinical suspicion should be raised upon findings of coexisting nodules, patchy consolidations, and cavitations, particularly in the presence of extrapulmonary infections such as those of the brain and subcutaneous tissues. A significant incidence of cavitations may be observed among immunosuppressed patients.

Radiomics Analysis of Multiphasic Computed Tomography Images for Distinguishing High-Risk Thymic Epithelial Tumors From Low-Risk Thymic Epithelial Tumors.

OBJECTIVES: The objective of this study is to preoperatively investigate the value of multiphasic contrast-enhanced computed tomography (CT)-based radiomics signatures for distinguishing high-risk thymic epithelial tumors (HTET) from low-risk thymic epithelial tumors (LTET) compared with conventional CT signatures. MATERIALS AND METHODS: Pathologically confirmed 305 thymic epithelial tumors (TETs), including 147 LTET (Type A/AB/B1) and 158 HTET (Type B2/B3/C), were retrospectively analyzed, and were randomly divided into training (n = 214) and validation cohorts (n = 91). All patients underwent nonenhanced, arterial contrast-enhanced, and venous contrast-enhanced CT analysis. The least absolute shrinkage and selection operator regression with 10-fold cross-validation was performed for radiomic models building, and multivariate logistic regression analysis was performed for radiological and combined models building. The performance of the model was evaluated by the area under the receiver operating characteristic curve (AUC of ROC), and the AUCs were compared using the Delong test. Decision curve analysis was used to evaluate the clinical value of each model. Nomogram and calibration curves were plotted for the combined model. RESULTS: The AUCs for radiological model in the training and validation cohorts were 0.756 and 0.733, respectively. For nonenhanced, arterial contrast-enhanced, venous contrast-enhanced CT and 3-phase images combined radiomics models, the AUCs were 0.940, 0.946, 0.960, and 0.986, respectively, in the training cohort, whereas 0.859, 0.876, 0.930, and 0.923, respectively, in the validation cohort. The combined model, including CT morphology and radiomics signature, showed AUCs of 0.990 and 0.943 in the training and validation cohorts, respectively. Delong test and decision curve analysis showed that the predictive performance and clinical value of the 4 radiomics models and combined model were greater than the radiological model ( P < 0.05). CONCLUSIONS: The combined model, including CT morphology and radiomics signature, greatly improved the predictive performance for distinguishing HTET from LTET. Radiomics texture analysis can be used as a noninvasive method for preoperative prediction of the pathological subtypes of TET.

Deep multiple instance learning for predicting chemotherapy response in non-small cell lung cancer using pretreatment CT images.

The individual prognosis of chemotherapy is quite different in non-small cell lung cancer (NSCLC). There is an urgent need to precisely predict and assess the treatment response. To develop a deep multiple-instance learning (DMIL) based model for predicting chemotherapy response in NSCLC in pretreatment CT images. Two datasets of NSCLC patients treated with chemotherapy as the first-line treatment were collected from two hospitals. Dataset 1 (163 response and 138 nonresponse) was used to train, validate, and test the DMIL model and dataset 2 (22 response and 20 nonresponse) was used as the external validation cohort. Five backbone networks in the feature extraction module and three pooling methods were compared. The DMIL with a pre-trained VGG16 backbone and an attention mechanism pooling performed the best, with an accuracy of 0.883 and area under the curve (AUC) of 0.982 on Dataset 1. While using max pooling and convolutional pooling, the AUC was 0.958 and 0.931, respectively. In Dataset 2, the best DMIL model produced an accuracy of 0.833 and AUC of 0.940. Deep learning models based on the MIL can predict chemotherapy response in NSCLC using pretreatment CT images and the pre-trained VGG16 with attention mechanism pooling yielded better predictions.

Predicting chemotherapy response in non-small-cell lung cancer via computed tomography radiomic features: Peritumoral, intratumoral, or combined?

Purpose: This study aims to evaluate the ability of peritumoral, intratumoral, or combined computed tomography (CT) radiomic features to predict chemotherapy response in non-small cell lung cancer (NSCLC). Methods: After excluding subjects with incomplete data or other types of treatments, 272 (Dataset 1) and 43 (Dataset 2, external validation) NSCLC patients who were only treated with chemotherapy as the first-line treatment were enrolled between 2015 and 2019. All patients were divided into response and nonresponse based on the response evaluation criteria in solid tumors, version 1.1. By using 3D slicer and morphological operations in python, the intra- and peritumoral regions of lung tumors were segmented from pre-treatment CT images (unenhanced) and confirmed by two experienced radiologists. Then radiomic features (the first order, texture, shape, et al.) were extracted from the above regions of interest. The models were trained and tested in Dataset 1 and further validated in Dataset 2. The performance of models was compared using the area under curve (AUC), confusion matrix, accuracy, precision, recall, and F1-score. Results: The radiomic model using features from the peritumoral region of 0-3 mm outperformed that using features from 3-6, 6-9, 9-12 mm peritumoral region, and intratumoral region (AUC: 0.95 versus 0.87, 0.86, 0.85, and 0.88). By the fusion of features from 0-3 and 3-6 mm peritumoral regions, the logistic regression model achieved the best performance, with an AUC of 0.97. This model achieved an AUC of 0.85 in the external cohort. Moreover, among the 20 selected features, seven features differed significantly between the two groups (p < 0.05). Conclusions: CT radiomic features from both the peri- and intratumoral regions can predict chemotherapy response in NSCLC using machine learning models. Combined features from two peritumoral regions yielded better predictions.

Multi-scale segmentation squeeze-and-excitation UNet with conditional random field for segmenting lung tumor from CT images.

BACKGROUND AND OBJECTIVE: Lung cancer counts among diseases with the highest global morbidity and mortality rates. The automatic segmentation of lung tumors from CT images is of vast significance. However, the segmentation faces several challenges, including variable shapes and different sizes, as well as complicated surrounding tissues. METHODS: We propose a multi-scale segmentation squeeze-and-excitation UNet with a conditional random field (M-SegSEUNet-CRF) to automatically segment lung tumors from CT images. M-SegSEUNet-CRF employs a multi-scale strategy to solve the problem of variable tumor size. Through the spatially adaptive attention mechanism, the segmentation SE blocks embedded in 3D UNet are utilized to highlight tumor regions. The dense connected CRF framework is further added to delineate tumor boundaries at a detailed level. In total, 759 CT scans of patients with lung cancer were used to train and evaluate the M-SegSEUNet-CRF model (456 for training, 152 for validation, and 151 for test). Meanwhile, the public NSCLC-Radiomics and LIDC datasets have been utilized to validate the generalization of the proposed method. The role of different modules in the M-SegSEUNet-CRF model is analyzed by the ablation experiments, and the performance is compared with that of UNet, its variants and other state-of-the-art models. RESULTS: M-SegSEUNet-CRF can achieve a Dice coefficient of 0.851 ± 0.071, intersection over union (IoU) of 0.747 ± 0.102, sensitivity of 0.827 ± 0.108, and positive predictive value (PPV) of 0.900 ± 0.107. Without a multi-scale strategy, the Dice coefficient drops to 0.820 ± 0.115; without CRF, it drops to 0.842 ± 0.082, and without both, it drops to 0.806 ± 0.120. M-SegSEUNet-CRF presented a higher Dice coefficient than 3D UNet (0.782 ± 0.115) and its variants (ResUNet, 0.797 ± 0.132; DenseUNet, 0.792 ± 0.111, and UNETR, 0.794 ± 0.130). Although the performance slightly declines with the decrease in tumor volume, M-SegSEUNet-CRF exhibits more obvious advantages than the other comparative models. CONCLUSIONS: Our M-SegSEUNet-CRF model improves the segmentation ability of UNet through the multi-scale strategy and spatially adaptive attention mechanism. The CRF enables a more precise delineation of tumor boundaries. The M-SegSEUNet-CRF model integrates these characteristics and demonstrates outstanding performance in the task of lung tumor segmentation. It can furthermore be extended to deal with other segmentation problems in the medical imaging field.

The Assessment of Prostate Cancer Aggressiveness Using a Combination of Quantitative Diffusion-Weighted Imaging and Dynamic Contrast-Enhanced Magnetic Resonance Imaging.

OBJECTIVE: To explore the value of combining dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) quantitative parameters with apparent diffusion coefficient (ADC) values in the diagnosis of prostate cancer. METHODS: The clinical data of 146 patients with prostate lesions, including 87 patients with prostate cancer (PCa) and 59 with benign prostatic hyperplasia (BPH), were collected. After DCE-MRI and diffusion-weighted imaging (DWI) prostate scans, the magnitude of the DCE-MRI transfer constant (Ktrans ), rate constant (kep ), the volume of the extravascular extracellular space (ve ), and the ADC between the groups were compared, and the correlations between the DCE-MRI parameters and Gleason scores were analyzed. The diagnostic efficacy of these quantitative parameters was assessed by the area under the receiver operating characteristic (ROC) curve. RESULTS: The DCE-MRI parameters Ktrans and kep were significantly greater in the PCa group than in the BPH group (p < 0.05). The ROC curve showed the area under the Ktrans, kep , and ADC curves to be 0.665, 0.658, and 0.782, respectively. When all three quantitative indicators were combined, the area under the ROC curve was 0.904, with sensitivity and specificity rates of 83.6% and 93.7%, respectively. The Gleason scores were positively correlated with the Ktrans, kep , and ve (r = 0.39, 0.572, 0.30, respectively; p < 0.05) and negatively correlated with the ADC (r = -0.525; p < 0.05). CONCLUSION: The DCE-MRI quantitative parameters Ktrans and kep , as well as the ADC value, provided effective references for the differential diagnosis of PCa and BPH, as well as more precise and reliable quantitative parameters for grading the aggressiveness of PCa.

An antibiotic stewardship programme to reduce inappropriate antibiotic prescribing for acute respiratory infections in rural Chinese primary care facilities: study protocol for a clustered randomised controlled trial.

BACKGROUND: Inappropriate prescribing of antibiotics for acute respiratory infections at the primary care level represents the major source of antibiotic misuse in healthcare, and is a major driver for antimicrobial resistance worldwide. In this study we will develop, pilot and evaluate the effectiveness of a comprehensive antibiotic stewardship programme in China's primary care hospitals to reduce inappropriate prescribing of antibiotics for acute respiratory infections among all ages. METHODS: We will use a parallel-group, cluster-randomised, controlled, superiority trial with blinded outcome evaluation but unblinded treatment (providers and patients). We will randomise 34 primary care hospitals from two counties within Guangdong province into the intervention and control arm (1:1 overall ratio) stratified by county (8:9 within-county ratio). In the control arm, antibiotic prescribing and management will continue through usual care. In the intervention arm, we will implement an antibiotic stewardship programme targeting family physicians and patients/caregivers. The family physician components include: (1) training using new operational guidelines, (2) improved management and peer-review of antibiotic prescribing, (3) improved electronic medical records and smart phone app facilitation. The patient/caregiver component involves patient education via family physicians, leaflets and videos. The primary outcome is the proportion of prescriptions for acute respiratory infections (excluding pneumonia) that contain any antibiotic(s). Secondary outcomes will address how frequently specific classes of antibiotics are prescribed, how frequently key non-antibiotic alternatives are prescribed and the costs of consultations. We will conduct a qualitative process evaluation to explore operational questions regarding acceptability, cultural appropriateness and burden of technology use, as well as a cost-effectiveness analysis and a long-term benefit evaluation. The duration of the intervention will be 12 months, with another 24 months' post-trial long-term follow-up. DISCUSSION: Our study is one of the first trials to evaluate the effect of an antibiotic stewardship programme in primary care settings in a low- or middle-income country (LMIC). All interventional activities will be designed to be embedded into routine primary care with strong local ownership. Through the trial we intend to impact on clinical practice and national policy in antibiotic prescription for primary care facilities in rural China and other LMICs. TRIAL REGISTRATION: ISRCTN, ID: ISRCTN96892547. Registered on 18 August 2019.

Deep CNN Model Using CT Radiomics Feature Mapping Recognizes EGFR Gene Mutation Status of Lung Adenocarcinoma.

To recognize the epidermal growth factor receptor (EGFR) gene mutation status in lung adenocarcinoma (LADC) has become a prerequisite of deciding whether EGFR-tyrosine kinase inhibitor (EGFR-TKI) medicine can be used. Polymerase chain reaction assay or gene sequencing is for measuring EGFR status, however, the tissue samples by surgery or biopsy are required. We propose to develop deep learning models to recognize EGFR status by using radiomics features extracted from non-invasive CT images. Preoperative CT images, EGFR mutation status and clinical data have been collected in a cohort of 709 patients (the primary cohort) and an independent cohort of 205 patients. After 1,037 CT-based radiomics features are extracted from each lesion region, 784 discriminative features are selected for analysis and construct a feature mapping. One Squeeze-and-Excitation (SE) Convolutional Neural Network (SE-CNN) has been designed and trained to recognize EGFR status from the radiomics feature mapping. SE-CNN model is trained and validated by using 638 patients from the primary cohort, tested by using the rest 71 patients (the internal test cohort), and further tested by using the independent 205 patients (the external test cohort). Furthermore, SE-CNN model is compared with machine learning (ML) models using radiomics features, clinical features, and both features. EGFR(-) patients show the smaller age, higher odds of female, larger lesion volumes, and lower odds of subtype of acinar predominant adenocarcinoma (APA), compared with EGFR(+). The most discriminative features are for texture (614, 78.3%) and the features of first order of intensity (158, 20.1%) and the shape features (12, 1.5%) follow. SE-CNN model can recognize EGFR mutation status with an AUC of 0.910 and 0.841 for the internal and external test cohorts, respectively. It outperforms the CNN model without SE, the fine-tuned VGG16 and VGG19, three ML models, and the state-of-art models. Utilizing radiomics feature mapping extracted from non-invasive CT images, SE-CNN can precisely recognize EGFR mutation status of LADC patients. The proposed method combining radiomics features and deep leaning is superior to ML methods and can be expanded to other medical applications. The proposed SE-CNN model may help make decision on usage of EGFR-TKI medicine.

Etiological distribution and clinical features of fever of unknown origin with pulmonary lesions in South China.

INTRODUCTION: Fever of unknown origin (FUO) with pulmonary lesions is a common syndrome in respiratory diseases, which can be caused by infection, cancer, connective tissue disease and other rare diseases of South China. In patients with FUO associated with pulmonary lesions, pathogeny should be identified and followed by an appropriate treatment strategy. OBJECTIVE: This study aimed to investigate the etiological distribution and clinical features of FUO with pulmonary lesions and to analyze the efficiency of different diagnostic methods for certain disease categories. METHODS: Patients hospitalized at the Guangzhou Institute of Respiratory Health from July 2012 to December 2016 who had fever ≥38.3°C that lasted ≥21 days, in whom the chest X-ray or computed tomography (CT) revealed definite pulmonary infiltration, and for whom, despite being examined for a week, no definitive diagnosis could be made, were considered for this study. RESULTS: A total of 104 patients were identified as having FUO with lung lesions, and 89.4% (93/104) patients were definitively diagnosed. Infectious disease was the most common cause (46.2%, 48/104). Histopathology was instrumental in the diagnosis of the causes of FUO with pulmonary manifestations, 47.3% (44/93) patients were diagnosed through histopathology, 35.4% (17/48) with infectious disease and 55.2% (16/29) with connective tissue diseases as the etiology were definitely diagnosed using histopathology. CONCLUSION: Most FUO with pulmonary lesions are identified during infections and autoimmune diseases. The most important diagnostic measure for FUO with pulmonary lesions is histopathology. Additionally, lung biopsy must be encouraged in multi-level hospitals in the future.

Computed Tomography-Based Radiomics Signature: A Potential Indicator of Epidermal Growth Factor Receptor Mutation in Pulmonary Adenocarcinoma Appearing as a Subsolid Nodule.

BACKGROUND: Lung adenocarcinoma (LADC) with epidermal growth factor receptor (EGFR) mutation is considered a subgroup of lung cancer sensitive to EGFR-targeted tyrosine kinase inhibitors. We aimed to develop and validate a computed tomography (CT)-based radiomics signature for prediction of EGFR mutation status in LADC appearing as a subsolid nodule. MATERIALS AND METHODS: A total of 467 eligible patients were divided into training and validation cohorts (n = 306 and 161, respectively). Radiomics features were extracted from unenhanced CT images by using Pyradiomics. A CT-based radiomics signature for distinguishing EGFR mutation status was constructed using the random forest (RF) method in the training cohort and then tested in the validation cohort. A combination of the radiomics signature with a clinical factors model was also constructed using the RF method. The performance of the model was evaluated using the area under the curve (AUC) of a receiver operating characteristic curve. RESULTS: In this study, 64.2% (300/467) of the patients showed EGFR mutations. L858R mutation of exon 21 was the most common mutation type (185/301). We identified a CT-based radiomics signature that successfully discriminated between EGFR positive and EGFR negative in the training cohort (AUC = 0.831) and the validation cohort (AUC = 0.789). The radiomics signature combined with the clinical factors model was not superior to the simple radiomics signature in the two cohorts (p > .05). CONCLUSION: As a noninvasive method, the CT-based radiomics signature can be used to predict the EGFR mutation status of LADC appearing as a subsolid nodule. IMPLICATIONS FOR PRACTICE: Lung adenocarcinoma (LADC) with epidermal growth factor receptor (EGFR) mutation is considered a subgroup of lung cancer that is sensitive to EGFR-targeted tyrosine kinase inhibitors. However, some patients with inoperable subsolid LADC are unable to undergo tissue sampling by biopsy for molecular analysis in clinical practice. A computed tomography-based radiomics signature may serve as a noninvasive biomarker to predict the EGFR mutation status of subsolid LADCs when mutational profiling is not available or possible.

Uncontrolled asthma phenotypes defined from parameters using quantitative CT analysis.

OBJECTIVE: Asthma is a heterogeneous disease with diverse clinical phenotypes that have been identified via cluster analyses. However, the classification of phenotypes based on quantitative CT (qCT) is poorly understood. The study was conducted to investigate CT determination of uncontrolled asthma phenotypes. METHODS: Sixty-five patients with uncontrolled asthma (37 with severe asthma, 28 with non-severe asthma) underwent detailed clinical, laboratory, and pulmonary function tests, as well as qCT analysis. Twenty-five healthy subjects were also included in this study and underwent clinical physical examinations, pulmonary function tests, and low-dose CT scans. RESULTS: The mean lumen area/body surface area ratio was smaller in patients with severe uncontrolled asthma compared with that in healthy subjects (9.84 mm2 [SD, 2.57 mm2], 11.96 mm2 [SD, 3.09 mm2]; p = 0.026). However, the percentage of mean wall area (WA) was greater (64.39% [SD, 2.55%], 62.09% [SD, 3.81%], p = 0.011). Air trapping (measured based on mean lung density and VI-856 [%] on expiratory scan) was greater in patients with severe uncontrolled asthma than in those with non-severe uncontrolled asthma and was higher in all patients with uncontrolled asthma than that in healthy subjects (all p < 0.001). Three CT-determined uncontrolled asthma phenotypes were identified. Cluster 1 had mild air trapping with or without proximal airway remodeling. Cluster 2 had moderate air trapping with or without proximal airway remodeling. Cluster 3 had severe air trapping with proximal airway remodeling. CONCLUSIONS: There was obvious air trapping and proximal airway remodeling in patients with severe uncontrolled asthma. The three CT-determined uncontrolled asthma phenotypes might reflect underlying mechanisms of disease in patient stratification and in the different stages of disease development. KEY POINTS: • Obvious air trapping and proximal airway remodeling were present in patients with severe uncontrolled asthma. • CT air trapping indices showed a good correlation with disease duration, total IgE, atopy, and OCS and ICS doses, and were even more strongly correlated with clinical lung function. • Three CT-determined uncontrolled asthma phenotypes were identified, which might reflect underlying mechanisms of disease in patient stratification and in the different stages of disease development.

CT-based radiomics signature for differentiating solitary granulomatous nodules from solid lung adenocarcinoma.

OBJECTIVES: Pulmonary granulomatous nodule (GN) with spiculated or lobulated appearance are indistinguishable from solid lung adenocarcinoma (SADC) based on CT morphological features, and partial false-positive findings on PET/CT. The objective of this study was to investigate the ability of quantitative CT radiomics for preoperatively differentiating solitary atypical GN from SADC. METHODS: 302 eligible patients (SADC = 209, GN = 93) were evaluated in this retrospective study and were divided into training (n = 211) and validation cohorts (n = 91). Radiomics features were extracted from plain and vein-phase CT images. The L1 regularized logistic regression model was used to identify the optimal radiomics features for construction of a radiomics model in differentiate solitary GN from SADC. The performance of the constructed radiomics model was evaluated using the area under curve (AUC) of receiver operating characteristic curve (ROC). RESULTS: 16.7% (35/209) of SADC were misdiagnosed as GN and 24.7% (23/93) of GN were misdiagnosed as lung cancer before surgery. The AUCs of combined radiomics and clinical risk factors were 0.935, 0.902, and 0.923 in the training cohort of plain radiomics(PR), vein radiomics, and plain and vein radiomics, and were 0.817, 0835, and 0.841 in the validation cohort of three models, respectively. PR combined with clinical risk factors (PRC) performed better than simple radiomics models (p < 0.05). The diagnostic accuracy of PRC in the total cohorts was similar to our radiologists (p ≥ 0.05). CONCLUSIONS: As a noninvasive method, PRC has the ability to identify SADC and GN with spiculation or lobulation.

Correlations between computed tomography and positron emission tomography/computed tomography findings and pathology in 6 cases of pulmonary epithelioid angiosarcoma.

Previous studies on primary pulmonary epithelioid angiosarcoma (PEA) have been mostly clinical or pathological case reports. We here summarize findings from computed tomography (CT) and positron emission tomography/computed tomography (PET/CT) analyses of PEA to improve the diagnosis and differentiation of this rare tumor.We conducted a retrospective analysis of the clinical findings, radiological imaging, and pathological findings of 6 cases of primary PEA confirmed by surgery, biopsy, and pathology. All cases were evaluated by CT and x-ray prior to surgery, and 2 cases were further examined by PET/CT.CT images indicated maximum tumor diameters of 2.4 to 9.8 cm and inhomogeneous density, with 1 case exhibiting nodular calcification. Contrast-enhanced CT revealed inhomogeneous enhancement with visible necrosis in all 6 cases, while 3 cases had hilar and mediastinal lymph node metastasis. Five cases displayed extensive tumor involvement with extension into the chest wall, mild-to-moderate levels of pleural effusion, and varying degrees of volume loss in the corresponding hemithorax. One case had limited pleural thickening and invasion. Preoperative PET/CT of 1 case revealed abnormal fluorine-18 fluorodeoxyglucose (F-FDG) uptake by the tumor and multiple enlarged right hilar and mediastinal lymph nodes, right diffuse pleural thickening, and systemic multiple bone metastasis. In the other case, PET/CT scan at 7 months after surgery revealed pleural thickening and mediastinal lymph nodes with increased F-FDG uptake on the surgical side. Immunohistochemistry analyses determined that all 6 tumors were positive for CD34, CD31, ERG, and vimentin.CT and PET/CT findings reveal that malignant characteristics, including extensive pleural thickening, invasion and metastasis, and pleural effusion, are common in PEA. Imaging data are only supportive; therefore, the final diagnosis should be based on pathology and immunohistochemistry analyses.

Is a 5-mm diameter an appropriate cut-off value for the diagnosis of atypical adenomatous hyperplasia and adenocarcinoma in situ on chest computed tomography and pathological examination?

BACKGROUND: Preinvasive lesions, such as atypical adenomatous hyperplasia (AAH) and adenocarcinoma in situ (AIS), usually appear as pure ground-glass nodules (pGGNs) on thin-section computed tomography (TSCT). AAH is usually less than 5 mm wide on imaging and pathological examinations. We aimed to determine whether a 5-mm cut-off value was appropriate for the diagnosis of AAH and AIS. METHODS: We retrospectively analyzed the performance of TSCT in evaluating 80 pathologically confirmed preinvasive lesions (33 AAH lesions in 31 patients and 47 AIS lesions in 45 patients). We compared the following characteristics between the AAH and AIS groups: lesion diameter, density, rim, lobulation, spiculation, vacuole sign, aerated bronchus sign, pleural indentation sign, and pathological findings. RESULTS: All 80 lesions appeared as pGGNs. On TSCT, the average diameter of AAH lesions (6.0±1.64 mm) was significantly smaller than that of AIS lesions (8.7±3.16 mm; P<0.001). The area under the curve (AUC) for diameter was 0.792, and the best diagnostic cut-off value was 6.99 mm. On gross pathological examination, the average diameter of AAH lesions (4.6±1.99 mm) was significantly smaller that of AIS lesions (6.8±2.06 mm; P<0.001). The AUC was 0.794, and the best diagnostic cut-off value was 4.5 mm. The vacuole sign was common in AIS (P=0.021). AAH did not significantly differ from AIS (P>0.05) in terms of average CT value, uniformity of density, morphology, rim, lobulation, spiculation, pleural indentation sign, and aerated bronchus sign. CONCLUSIONS: Lesion size and the vacuole sign were beneficial in the diagnosis of AAH and AIS. The vacuole sign was common in AIS. The best diagnostic cut-off value of nodular diameter for differentiating between AAH and AIS was 6.99 mm on TSCT and 4.5 mm on gross pathology.

Malignant solitary pulmonary nodules: assessment of mass growth rate and doubling time at follow-up CT.

BACKGROUND: The differentiation of benign and malignant solitary pulmonary nodules (SPNs), especially subsolid nodules, is still challenging because of the small size, slow growth, and atypical imaging characteristics of these nodules. We aimed to determine the significance of mass growth rate (MGR) and mass doubling time (MDT) at follow-up CT of malignant SPNs. METHODS: This retrospective study included 167 patients (169 SPNs, diameter 8-30 mm). Among the 169 SPNs, 114 malignant SPNs were classified into three types: pure ground-glass nodules (pGGNs), part-solid nodules (pSNs), and solid nodules (SNs). These patients were followed up for at least 3 months. Three-dimensional manual segmentation was performed for all these nodules, and the intra- and inter-observer variabilities of diameter, volume, and mass measurement were assessed. From initial and follow-up CT scans, growth rates of the diameter, volume, and mass of the SPNs were compared. MDT and volume doubling time (VDT) were calculated and were compared among groups. RESULTS: Mass measurements had the best inter-observer consistency and intra-observer repeatability; the coefficients of variation of the mass measurements were the smallest. The mean growth rates of the diameter, volume, and mass of pGGNs, pSNs, and SNs significantly differed at different time points (P<0.001). Mean MDTs and VDTs of pGGNs, pSNs, and SNs were 655 vs. 848 days, 462 vs. 598 days, and 230 vs. 267 days, respectively (P<0.05). CONCLUSIONS: Mass measurements are an objective and accurate indicator in SPN assessment. During a 2-year follow-up, the mean growth rates of the diameter, volume, and mass of pGGNs, pSNs, and SNs differed at different time points, the greatest difference was observed in mean MGR. Mean MDT of malignant SPNs is less than the mean VDT.

A new approach to predict lymph node metastasis in solid lung adenocarcinoma: a radiomics nomogram.

BACKGROUND: Lymph node metastasis (LNM) of lung cancer is an important factor related to survival and recurrence. The association between radiomics features of lung cancer and LNM remains unclear. We developed and validated a radiomics nomogram to predict LNM in solid lung adenocarcinoma. METHODS: A total of 159 eligible patients with solid lung adenocarcinoma were divided into training (n=106) and validation cohorts (n=53). Radiomics features were extracted from venous-phase CT images. We built a radiomics nomogram using a multivariate logistic regression model combined with CT-reported lymph node (LN) status. The performance of the radiomics nomogram was evaluated using the area under curve (AUC) of receiver operating characteristic curve. We performed decision curve analysis (DCA) within training and validation cohorts to assess the clinical usefulness of the nomogram. RESULTS: Fourteen radiomics features were chosen from 94 candidate features to build a radiomics signature that significantly correlated with LNM. The model showed good calibration and discrimination in the training cohort, with an AUC of 0.871 (95% CI: 0.804-0.937), sensitivity of 85.71% and specificity of 77.19%. In the validation cohort, AUC was 0.856 (95% CI: 0.745-0.966), sensitivity was 91.66%, and specificity was 82.14%. DCA demonstrated that the nomogram was clinically useful. The nomogram also showed good predictive ability in patients at high risk for LNM in the CT-reported LN negative (cN0) subgroup. CONCLUSIONS: The radiomics nomogram, based on preoperative CT images, can be used as a noninvasive method to predict LNM in patients with solid lung adenocarcinoma.

Erlotinib intercalating pemetrexed/cisplatin versus erlotinib alone in Chinese patients with brain metastases from lung adenocarcinoma: a prospective, non-randomised, concurrent controlled trial (NCT01578668).

OBJECTIVE: Erlotinib has a synergistic effect with pemetrexed for treating non-squamous non-small-cell lung cancer. We investigated the efficacy and safety of erlotinib (E) in combination with pemetrexed/cisplatin (E-P) in Chinese patients with lung adenocarcinoma with brain metastases. DESIGN: Patients who were erlotinib-naïve or pemetrexed-naïve were assigned in parallel to receive either E or E-P. The primary endpoint was the intracranial overall response rate (ORRi). RESULTS: Sixty-nine patients with lung adenocarcinoma with brain metastases received E (n=35) or E-P (n=34) from January 2012 to November 2014. Demographics and patient characteristics were well balanced between the two groups, including epidermal growth factor receptor (EGFR) status, sex, age, smoking status, Eastern Cooperative Oncology Group (ECOG) performance status, brain metastases and number of prior treatments. ORRi in the E-P arm was superior to that in the E arm (79% vs 48%, p=0.008). Compared with E as the first-line treatment, E-P was associated with better intracranial progression-free survival (PFSi, median: 9 vs 2 months, p=0.027) and systemic PFS (median: 8 vs 2 months, p=0.006). The most frequent E-related adverse events were higher in the combination arm. No new safety signals were detected. The side effects were tolerable, and there were no drug-related deaths. CONCLUSION: Our study suggests that the E-P combination may be effective in Chinese patients with lung adenocarcinoma with brain metastases, with improved PFS in treatment-naïve patients. Toxicities are tolerable, and there are more E-related side effects.

Quantitative Low-Dose Computed Tomography of the Lung Parenchyma and Airways for the Differentiation between Chronic Obstructive Pulmonary Disease and Asthma Patients.

BACKGROUND: It is difficult to differentiate between chronic obstructive pulmonary disease (COPD) and asthma in clinics; therefore, for diagnostic purposes, imaging-based measurements could be beneficial to differentiate between the two diseases. OBJECTIVES: We aim to analyze quantitative measurements of the lung and bronchial parameters that are provided by low-dose computed tomography (CT) to differentiate COPD and asthma from an imaging perspective. MATERIALS AND METHODS: 69 COPD patients, 52 asthma patients, and 20 healthy subjects were recruited to participate in CT imaging and pulmonary function tests (PFTs). Comparative analysis was performed to identify differences between COPD and asthma in CT measurements. PFT measurements enabled validation of the differentiation between COPD and asthma patients. RESULTS: There were significant differences among the COPD, asthma, and healthy control groups. The differences were more significant among the following: inspiratory emphysema index (EI)-950 (%), expiratory lung volume, expiratory mean lung density (MLD), and expiratory EI-950 (%) and EI-850 (%). The COPD group had a significantly higher EI-950 (%) than the asthma group (p = 0.008). There were significant differences among the three groups in lumen area (LA), wall area (WA), total area, and Pi10WA. The asthma group had significantly higher WA%/WV% than both the COPD (p = 0.002) and the control group (p = 0.012). There was high sensitivity in EI-950 (%), EI-850 (%) and expiratory MLD in the parenchyma and high sensitivity in LA and Pi10WA in small airways in the differential diagnosis of COPD and asthma. CONCLUSION: To aid the diagnosis, CT can provide quantitative measurements to differentiate between COPD and asthma patients.