Nomograms integrating CT radiomic and deep learning signatures to predict overall survival and progression-free survival in NSCLC patients treated with chemotherapy.

OBJECTIVES: This study aims to establish nomograms to accurately predict the overall survival (OS) and progression-free survival (PFS) in patients with non-small cell lung cancer (NSCLC) who received chemotherapy alone as the first-line treatment. MATERIALS AND METHODS: In a training cohort of 121 NSCLC patients, radiomic features were extracted, selected from intra- and peri-tumoral regions, and used to build signatures (S1 and S2) using a Cox regression model. Deep learning features were obtained from three convolutional neural networks and utilized to build signatures (S3, S4, and S5) that were stratified into over- and under-expression subgroups for survival risk using X-tile. After univariate and multivariate Cox regression analyses, a nomogram incorporating the tumor, node, and metastasis (TNM) stages, radiomic signature, and deep learning signature was established to predict OS and PFS, respectively. The performance was validated using an independent cohort (61 patients). RESULTS: TNM stages, S2 and S3 were identified as the significant prognosis factors for both OS and PFS; S2 (OS: (HR (95%), 2.26 (1.40-3.67); PFS: (HR (95%), 2.23 (1.36-3.65)) demonstrated the best ability in discriminating patients with over- and under-expression. For the OS nomogram, the C-index (95% CI) was 0.74 (0.70-0.79) and 0.72 (0.67-0.78) in the training and validation cohorts, respectively; for the PFS nomogram, the C-index (95% CI) was 0.71 (0.68-0.81) and 0.72 (0.66-0.79). The calibration curves for the 3- and 5-year OS and PFS were in acceptable agreement between the predicted and observed survival. The established nomogram presented a higher overall net benefit than the TNM stage for predicting both OS and PFS. CONCLUSION: By integrating the TNM stage, CT radiomic signature, and deep learning signatures, the established nomograms can predict the individual prognosis of NSCLC patients who received chemotherapy. The integrated nomogram has the potential to improve the individualized treatment and precise management of NSCLC patients.

Multi-site, Multi-domain Airway Tree Modeling.

Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to the quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and extensive clinical efforts for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Both quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage (https://atm22.grand-challenge.org/).

Two-stage contextual transformer-based convolutional neural network for airway extraction from CT images.

Accurate airway segmentation from computed tomography (CT) images is critical for planning navigation bronchoscopy and realizing a quantitative assessment of airway-related chronic obstructive pulmonary disease (COPD). Existing methods face difficulty in airway segmentation, particularly for the small branches of the airway. These difficulties arise due to the constraints of limited labeling and failure to meet clinical use requirements in COPD. We propose a two-stage framework with a novel 3D contextual transformer for segmenting the overall airway and small airway branches using CT images. The method consists of two training stages sharing the same modified 3D U-Net network. The novel 3D contextual transformer block is integrated into both the encoder and decoder path of the network to effectively capture contextual and long-range information. In the first training stage, the proposed network segments the overall airway with the overall airway mask. To improve the performance of the segmentation result, we generate the intrapulmonary airway branch label, and train the network to focus on producing small airway branches in the second training stage. Extensive experiments were performed on in-house and multiple public datasets. Quantitative and qualitative analyses demonstrate that our proposed method extracts significantly more branches and longer lengths of the airway tree while accomplishing state-of-the-art airway segmentation performance. The code is available at https://github.com/zhaozsq/airway_segmentation.

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.

Using CT radiomic features based on machine learning models to subtype adrenal adenoma.

BACKGROUND: Functioning and non-functioning adrenocortical adenoma are two subtypes of benign adrenal adenoma, and their differential diagnosis is crucial. Current diagnostic procedures use an invasive method, adrenal venous sampling, for endocrinologic assessment. METHODS: This study proposes establishing an accurate differential model for subtyping adrenal adenoma using computed tomography (CT) radiomic features and machine learning (ML) methods. Dataset 1 (289 patients with adrenal adenoma) was collected to develop the models, and Dataset 2 (54 patients) was utilized for external validation. Cuboids containing the lesion were cropped from the non-contrast, arterial, and venous phase CT images, and 1,967 features were extracted from each cuboid. Ten discriminative features were selected from each phase or the combined phases. Random forest, support vector machine, logistic regression (LR), Gradient Boosting Machine, and eXtreme Gradient Boosting were used to establish prediction models. RESULTS: The highest accuracies were 72.7%, 72.7%, and 76.1% in the arterial, venous, and non-contrast phases, respectively, when using radiomic features alone with the ML classifier of LR. When features from the three CT phases were combined, the accuracy of LR reached 83.0%. After adding clinical information, the area under the receiver operating characteristic curve increased for all the machine learning methods except for LR. In Dataset 2, the accuracy of LR was the highest, reaching 77.8%. CONCLUSION: The radiomic features of the lesion in three-phase CT images can potentially suggest the functioning or non-functioning nature of adrenal adenoma. The resulting radiomic models can be a non-invasive, low-cost, and rapid method of minimizing unnecessary testing in asymptomatic patients with incidentally discovered adrenal adenoma.

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.

Predictive Radiomic Models for the Chemotherapy Response in Non-Small-Cell Lung Cancer based on Computerized-Tomography Images.

The heterogeneity and complexity of non-small cell lung cancer (NSCLC) tumors mean that NSCLC patients at the same stage can have different chemotherapy prognoses. Accurate predictive models could recognize NSCLC patients likely to respond to chemotherapy so that they can be given personalized and effective treatment. We propose to identify predictive imaging biomarkers from pre-treatment CT images and construct a radiomic model that can predict the chemotherapy response in NSCLC. This single-center cohort study included 280 NSCLC patients who received first-line chemotherapy treatment. Non-contrast CT images were taken before and after the chemotherapy, and clinical information were collected. Based on the Response Evaluation Criteria in Solid Tumors and clinical criteria, the responses were classified into two categories: response (n = 145) and progression (n = 135), then all data were divided into two cohorts: training cohort (224 patients) and independent test cohort (56 patients). In total, 1629 features characterizing the tumor phenotype were extracted from a cube containing the tumor lesion cropped from the pre-chemotherapy CT images. After dimensionality reduction, predictive models of the chemotherapy response of NSCLC with different feature selection methods and different machine-learning classifiers (support vector machine, random forest, and logistic regression) were constructed. For the independent test cohort, the predictive model based on a random-forest classifier with 20 radiomic features achieved the best performance, with an accuracy of 85.7% and an area under the receiver operating characteristic curve of 0.941 (95% confidence interval, 0.898-0.982). Of the 20 selected features, four were first-order statistics of image intensity and the others were texture features. For nine features, there were significant differences between the response and progression groups (p < 0.001). In the response group, three features, indicating heterogeneity, were overrepresented and one feature indicating homogeneity was underrepresented. The proposed radiomic model with pre-chemotherapy CT features can predict the chemotherapy response of patients with non-small cell lung cancer. This radiomic model can help to stratify patients with NSCLC, thereby offering the prospect of better treatment.

Modification of association between prior lung disease and lung cancer by inhaled arsenic: A prospective occupational-based cohort study in Yunnan, China.

Arsenic and prior lung diseases have been shown to increase lung cancer risk; however, little is known about their joint effects. The aim of our study was to analyze the joint effects of inhaled arsenic and prior lung diseases on lung cancer risk within a occupational cohort. The interactions of prior lung diseases and inhaled arsenic were analyzed based on multiplicative and additive scales in the Cox proportional hazards model. Compared with low arsenic exposure and no history of asthma, the hazard ratios (HRs) of high arsenic exposure with asthma, high arsenic exposure without asthma and low arsenic exposure with asthma were 2.61 (95% CI: 1.71-4.00), 2.60 (95% CI: 1.93-3.51) and 2.49 (95% CI: 1.53-4.06), respectively. Based on the multiplicative scale in the Cox proportional hazards model, the HR of the interaction of asthma and arsenic on lung cancer risk was 0.45 (95% CI: 0.25-0.80). Based on the additive scale, the relative excess risk due to interaction between asthma and arsenic was -1.41 (95% CI: -2.81 to -0.02). Our study provides strong evidence that arsenic exposure is associated with lung cancer risk. A significant negative interaction between asthma and arsenic on lung cancer risk is observed.

[A cohort study on risk factors of lung cancer in Yunnan tin miners].

BACKGROUND: Smoking is a major cause of lung cancer. Studies of lung cancer among miners have shown that occupational exposure also played an important role. The aim of this study is to investigate radon, cigarette use and other risk factors of lung cancer in Yunnan tin miners and to provide a scientific basis for the prevention and control of occupational lung cancer. METHODS: A prospective cohort study was conducted among Yunnan tin miners, the associations between potential risk factors for lung cancer were analyzed by multivariate Cox regression model. Effects of age at first radon exposure and radon exposure rate on lung cancer risk were analyzed. The relationship between cumulative working level month and lung cancer was analyzed according to smoking status. The joint effect of tobacco use and cumulative radon exposure was analyzed based on additive and multiplicative models. RESULTS: Increased risk of lung cancer was associated with age at enrollment, tobacco use, prior bronchitis, and cumulative arsenic and radon exposure, while higher education level was associated with decreased lung cancer risk. An inverse effect of radon exposure rate was observed. There was no significant association between lung cancer risk and first radon exposure age. There was a significant additive interaction between tobacco use and radon exposure on lung cancer risk. CONCLUSIONS: Several risk factors may contribute to the high incidence of lung cancer in Yunnan tin miners. Further studies are warranted to evaluate joint effect of different risk factors.

[Analysis of sensitivity and specificity of sputum cytology screening for lung cancer with different positive criteria].

OBJECTIVE: To discuss the effect of different positive criteria on the sensitivity and specificity of sputum cytology screening for lung cancer among Yunnan tin miners. METHODS: 9223 Yunnan tin miners who received at least one annual sputum cytology screening for lung cancer during the period between 1992 and 1999 were recruited in the study. At time of enrollment, all participants were aged over 40 years old, had at least 10 years of employment as an underground miner and(or) smelter, and had not been diagnosed with malignancy. In our study, a true positive was categorized as having at least one prior positive sputum screening and a diagnosis of lung cancer, while a true negative, by our definition, signified negative sputum examinations and no diagnosis of lung cancer during the follow up time. Based on different positive criteria, sensitivity and specificity of sputum cytology were computed and receiver operating characteristic (ROC) curve analysis was conducted. Z statistic was used to test the differences of the area under ROC based on Hanley and McNeil method. RESULTS: By the end of following up on December 31, 2001, a total 500 lung cancer cases were diagnosed among 9223 participants: most were squamous cell carcinoma (55.8% (222/398)) and central lung cancers (68.5% (316/461)). 150 lung cancer cases had a previous positive sputum screening result. When positive criteria were taken as grave atypical metaplasia, moderate atypical metaplasia and slight atypical metaplasia, the corresponding sensitivities were 30.0% (150/500), 36.4% (182/500), 53.0% (265/500) respectively; while the corresponding specificities were 98.9% (8628/8723), 95.1% (8611/8723), 77.9% (7033/8723) respectively. The areas under ROC curve according to different positive criterias were 0.645 (95%CI: 0.635 - 0.654), 0.657 (95%CI: 0.668 - 0.667), 0.655 (95%CI: 0.645 - 0.664) respectively. There were no significant differences found in the comparisons between grave and moderate atypical metaplasia, grave and slight atypical metaplasia, moderate and slight atypical metaplasia(Z statistics were 0.780, 0.645, 0.209 respectively, all P values > 0.05). CONCLUSION: While the standard of positive criteria for diagnosis of lung cancer decreased, the sensitivity of sputum cytology screening increased and the specificity decreased. Since there were no significant differences of accuracy for different positive criteria.

Prior lung disease and lung cancer risk in an occupational-based cohort in Yunnan, China.

We used the data from a prospective cohort study among tin miners in Yunnan, China to investigate whether prior lung disease is a risk factor for lung cancer. Information on prior lung disease was obtained from baseline questionnaires. The Cox proportional hazards model was used to examine the relationship between prior lung disease and lung cancer risk. From 1992 to 2001, a total of 502 lung cancer cases were confirmed among 9295 cohort participants. Prior chronic bronchitis was associated with an increase in lung cancer risk with an adjusted HR of 1.50 (95% CI: 1.24-1.81). There was an increased risk of developing squamous cell carcinoma in the setting of prior chronic bronchitis and small cell carcinoma in association with asthma with an adjusted HRs of 1.57 (95% CI: 1.19-2.09) and 2.56 (95% CI: 1.38-4.75), respectively. This prospective study provides further evidence that prior chronic bronchitis correlates with increased lung cancer risk, especially for squamous cell carcinoma. Asthma is associated with increased risk of small cell lung carcinoma.

Association between sputum atypia and lung cancer risk in an occupational cohort in Yunnan, China.

BACKGROUND: Individuals with cytologic atypia in sputum may be at high risk for the development of lung cancer. METHODS: A prospective cohort study was conducted among occupational tin miners in Yunnan, China, based on an annual lung cancer screening program. Sputum samples were collected prospectively at baseline and the following seven annual screenings. The associations between risk factors and sputum cytology were analyzed by univariate and multivariate logistic regression. A proportional hazard model was used to analyze the association between the baseline sputum results and the incidence of lung cancer. The effect of consecutive sputum cytology on the increase of lung cancer risk was analyzed by logistic regression. RESULTS: Sputum cytologic atypia was associated with age, smoking, occupational radon and arsenic exposure, and asthma. Sputum cytologic atypia was an independent risk factor for lung cancer with an adjusted hazard ratio (HR) of 3.82 (95% confidence interval [CI], 2.82 to 5.18) in comparing normal to moderate or worse atypia. Compared to the lung cancer risk associated with normal sputum, the risk was significantly higher according to the degree of atypia for squamous carcinomas, small cell lung cancer and central lung cancer, with adjusted HRs of 5.70 (95% CI, 3.78 to 8.59), 3.32 (95% CI, 1.31 to 8.45), and 4.93 (95% CI, 3.51 to 6.92), respectively. CONCLUSIONS: Sputum atypia is associated with an increased risk of lung cancer. Sputum cytologic examination combined with other screening examinations may play an important role in the early detection of lung cancer or in the selection of the optimal target population for more intensive lung cancer screening among this occupational cohort or similar population.

[Retrospective analysis of screening results of lung cancer cases among occupational population at high risk of lung cancer].

BACKGROUND: Lung cancer has become the leading cause of the cancer death in China. Population-based lung cancer screening is still in controversy. The objective of this study is to analyze the effect of annual chest radiography and sputum cytological screening conducted in high lung cancer risk population who were exposed to work related carcinogens. METHODS: A retrospective analysis was conducted to evaluate the screening results of the lung cancer cases diagnosed from 1992 to 2001 in the miners of Yunnan tin mine. RESULTS: A total of 9317 miners had been screened annually from 1992 to 1999. A total of 46 779 chest radiography and 45 672 sputum cytological examinations had been conducted, and 793 cohort subjects had at least one positive result. The annual positive detection rate ranged from 1214.1/100 000 to 3482.7/100 000. By December 31, 2001, 433 lung cancer cases had been confirmed, 371 cases out of them had cytological/pathological evidence, and 55.0% were squamous cell carcinoma followed by adenocarcinoma and small cell carcinoma. Stage I or II accounted for 24%. 62.1% of the cases had at least one positive screening result, while 165 cases were detected by chest radiography alone, 56 were detected by sputum cytology, and 48 were detected by both screening modalities. 64.2% of X-ray detected cases were squamous/adenous carcinomas and 75.0% of cytological detected cases were squamous carcinoma. 80.8% of early stage cases had at least one previous positive finding from screening. CONCLUSIONS: Annual lung cancer screening with combination of chest radiography and sputum cytology play some extent role in early detection of lung cancer in high risk population. The results may provide some primary data for lung cancer screening in special population who are at high risk of lung cancer in China.

[Establishment of cohort to study lung cancer in Yunnan tin miners].

BACKGROUND: To establish a cohort for the study of risk factors of lung cancer, and to support the study of early biomarkers and prevention of lung cancer. METHODS: Designed a special population-based prospectively dynamic cohort among radon- and arsenic-exposed tin miners aged 40 or more years old with at least 10 years of occupational exposure in Yunnan Province, P.R.China. The mass screenings with sputum cytology and chest X-ray were conducted annually. The baseline information was collected for assessing demographic characteristics and risk factors. The multiple sputum specimens, chest radiographs and numerous biologic specimens have been collected and stored. RESULTS: From 1992-1999, 9143 miners have been enrolled and 460 new cases of lung cancer have been found. There had 47655 person-time chest radiographs and 46625 person-time sputum cytology among the cohort in 8 years. The relative risks of age-adjusted exposure to chronic bronchitis, silicosis, and tobacco were 1.73, 1.46, and 1.32 respectively. CONCLUSIONS: A cohort of unique occupationally-exposed tin miners with an extensive biologic specimen repository and data bank has been successfully established. Although occupational exposures are the predominant risk factors among the high risk miners, lung cancer risk is also associated with chronic obstructive lung disease (chronic bronchitis and silicosis) and a number of measures of exposure to tobacco smoke, including early age of first use, duration, and cumulative exposure.