CT-based radiomics and machine learning to predict spread through air space in lung adenocarcinoma

Jiang, Changsi; Luo, Yan; Yuan, Jialin; You, Shuyuan; Chen, Zhiqiang; Wu, Mingxiang; Wang, Guangsuo; Gong, Jingshan

doi:10.1007/s00330-020-06694-z

Cited by 52 publications

(53 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A number of studies have investigated different approaches to identify STAS in ADCs pre-and intraoperatively. 13,19,[28][29][30][31] To name just a few, Toyokawa et al reported that the presence of notch and the absence of ground-glass opacity (GGO) were associated significantly with STAS in ADCs. 29 13 Besides, it has been reported that intraoperative imprint cytology with the N-H classification for ADC was well correlated with the STAS status of the tumor.…”

Section: Discussionmentioning

confidence: 99%

Could tumor spread through air spaces benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-institutional study

et al. 2020

View full text Add to dashboard Cite

Background: The benefit of adjuvant chemotherapy (ACT) remains unknown for patients with stage I lung adenocarcinoma (ADC) with spread through air spaces (STAS). This study investigated the effect of adjuvant chemotherapy in stage I ADC/STAS-positive patients. Methods: A total of 3346 patients with stage I ADC from five institutions in China were identified from 2009 to 2013, of whom 1082 were diagnosed with STAS (32.3%). By using the Kaplan–Meier method and Cox proportional hazard regression model, we explored the impact of STAS on prognosis, and determined if the use of adjuvant chemotherapy was associated with improved outcomes in patients with stage I ADC/STAS-positive. A validation cohort was also included in this study. Results: Patients with stage I ADC/STAS-positive in the primary cohort had unfavorable overall survival (OS) and disease-free survival (DFS). A multivariate Cox regression model confirmed the survival disadvantages of STAS in patients with stage I ADC [OS: hazards ratio (HR) = 1.877, 95% confidence interval (CI): 1.579–2.231; p < 0.001; DFS: HR = 1.895, 95% CI: 1.614–2.225; p < 0.001]. Lobectomy was associated with better OS and DFS than sublobar resection (SR) in both stage IA and IB ADC/STAS-positive. Similar results were observed in the validation cohort. For patients with stage IB ADC/STAS-positive, ACT was revealed as an independent factor for favorable survival (OS: HR = 0.604, 95% CI: 0.397–0.919; p = 0.018; DFS: HR = 0.565, 95% CI: 0.372–0.858; p = 0.007). However, among patients with stage IA ADC/STAS-positive, ACT was associated with improved outcomes only for those undergoing SR (OS: HR = 0.787, 95% CI: 0.359–0.949; p = 0.034; DFS: HR = 0.703, 95% CI: 0.330–0.904; p = 0.029). Conclusion: The presence of STAS was correlated with poor prognosis in patients with stage I ADC. Our study suggested that ACT might be considered for patients with stage IB ADC/STAS-positive and those with stage IA ADC/STAS-positive who underwent SR.

show abstract

Section: Discussionmentioning

confidence: 99%

Could tumor spread through air spaces benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-institutional study

et al. 2020

View full text Add to dashboard Cite

show abstract

“…However, the problem of multicollinearity leading to predictive model overfitting cannot be ignored. Jiang et al (25) found that the random forest model built with reliable radiomics features (intraclass correlation coefficients ≥0.75) could obtain a good predictive performance in spreading through the air space in lung adenocarcinoma. Traverso et al (26) proposed that machine learning-based radiomics methods may be useful for robust predictive model development.…”

Section: Discussionmentioning

confidence: 99%

A machine learning-based prediction of the micropapillary/solid growth pattern in invasive lung adenocarcinoma with radiomics

He¹,

Song²,

Wang³

et al. 2021

Transl Lung Cancer Res

View full text Add to dashboard Cite

Background: Micropapillary/solid (MP/S) growth patterns of lung adenocarcinoma are vital for making clinical decisions regarding surgical intervention. This study aimed to predict the presence of a MP/S component in lung adenocarcinoma using radiomics analysis.Methods: Between January 2011 and December 2013, patients undergoing curative invasive lung adenocarcinoma resection were included. Using the "PyRadiomics" package, we extracted 90 radiomics features from the preoperative computed tomography (CT) images. Subsequently, four prediction models were built by utilizing conventional machine learning approaches fitting into radiomics analysis: a generalized linear model (GLM), Naïve Bayes, support vector machine (SVM), and random forest classifiers.The models' accuracy was assessed using a receiver operating curve (ROC) analysis, and the models' stability was validated both internally and externally.Results: A total of 268 patients were included as a primary cohort, and 36.6% (98/268) of them had lung adenocarcinoma with an MP/S component. Patients with an MP/S component had a higher rate of lymph node metastasis (18.4% versus 5.3%) and worse recurrence-free and overall survival. Five radiomics features were selected for model building, and in the internal validation, the four models achieved comparable performance of MP/S prediction in terms of area under the curve (AUC):

show abstract

“…Our results showed that the AUC of the nomogram model was slightly larger than that of the clinical model in both the training and test sets, but there were no statistically significant differences between the nomogram and clinical models in either the training set or test set (P = 0.2108 and 0.1324, respectively), which indicated that CT image features could provide plenty of information for making a preliminary judgment on the existence of STAS. Jiang et al developed a random forest model using CT-based radiomics features and achieved an AUC of 0.754 for predicting the existence of STAS [ 18 ]. Another study built a Naïve Bayes model using five radiomics features to predict STAS that showed good performance with an AUC of 0.63 in internal validation and an AUC of 0.69 in external validation [ 17 ].…”

Section: Discussionmentioning

confidence: 99%

“…Some studies on radiomics showed that radiomics features performed well for clinical decision making in patients with lung cancer [ 15 , 16 ]. To date, there have been two studies on the radiomics analysis of STAS, which predicted the existence of STAS by establishing different models [ 17 , 18 ]. Although the results showed that STAS could be predicted preoperatively, the methods and results of the two papers were not completely consistent.…”

Section: Introductionmentioning

confidence: 99%

Radiomics nomograms of tumors and peritumoral regions for the preoperative prediction of spread through air spaces in lung adenocarcinoma

Zhuo

Feng

Yang

et al. 2020

Translational Oncology

View full text Add to dashboard Cite

To evaluate the clinical features and radiomics nomograms of tumors and peritumoral regions for the preoperative prediction of the presence of spread through air spaces (STAS) in patients with lung adenocarcinoma. A total of 107 STAS-positive lung adenocarcinomas were selected and matched to 105 STAS-negative lung adenocarcinomas. Thin-slice CT imaging annotation and region of interest (ROI) segmentation were performed with semi-automatic in-house software. Radiomics features were extracted from all nodules and incremental distances of 5, 10, and 15 mm outside the lesion segmentation. A radiomics nomogram was established with multivariable logistic regression based on clinical and radiomics features. The maximum diameter of the solid component and mediastinal lymphadenectasis were selected as independent predictors of STAS. The radiomics nomogram of lung nodules showed especially good prediction in the training set [area under the curve (AUC), 0.98; 95% confidence interval (CI), 0.97–1.00] and test set (AUC, 0.99; 95% CI, 0.97–1.00). The radiomics nomogram of peritumoral regions also showed good prediction, but the fitting degrees of the calibration curves were not good. Our study may provide guidance for surgical methods in patients with lung adenocarcinoma.

show abstract

CT-based radiomics and machine learning to predict spread through air space in lung adenocarcinoma

Cited by 52 publications

References 18 publications

Could tumor spread through air spaces benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-institutional study

Could tumor spread through air spaces benefit from adjuvant chemotherapy in stage I lung adenocarcinoma? A multi-institutional study

A machine learning-based prediction of the micropapillary/solid growth pattern in invasive lung adenocarcinoma with radiomics

Radiomics nomograms of tumors and peritumoral regions for the preoperative prediction of spread through air spaces in lung adenocarcinoma

Contact Info

Product

Resources

About