3D convolutional neural network model from contrast-enhanced CT to predict spread through air spaces in non-small cell lung cancer

Tao, Junli; Liang, Changyu; Yin, Kai; Fang, Jiayang; Chen, Bohui; Wang, Zhenyu; Xiao, Lan; Zhang, Jiuquan

doi:10.1016/j.diii.2022.06.002

Cited by 15 publications

(7 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They achieved satisfactory performance with an AUC of 0.82 and an accuracy of 74%. Similar results were achieved by other studies, with accuracy ranging from 0.66 to 0.93 [37][38][39][40]. However, results seem strictly dependent on CT characteristics and scarcely reproducible, with only one study attempting to use radiomics tools in a heterogeneous dataset [39].…”

Section: Radiomics Modelsupporting

confidence: 80%

“…In recent years, several studies have focused on AI-based and radiomics models aimed at predicting STAS with promising results [35][36][37][38][39][40] (Table 2). To assess the value of radiomics in predicting STAS in stage I lung adenocarcinoma…”

Section: Spread Through Air Spacesmentioning

confidence: 99%

“…Unfortunately, we are far from a reliable preoperative prediction of STAS, which remains a histological characteristic tested postoperatively. Limitations in a future clinical transition lie in the heterogeneity of studies and datasets that make comparison and reproducibility very difficult [37][38][39][40]. Moreover, many studies are monocentric and do not include external validation, making their applicability questionable [39].…”

Section: Future Applications and Limitationsmentioning

confidence: 99%

See 2 more Smart Citations

Lung Cancer Surgery in Octogenarians: Implications and Advantages of Artificial Intelligence in the Preoperative Assessment

Bassi,

Vaz Sousa,

Zacchini

et al. 2024

Healthcare

View full text Add to dashboard Cite

The general world population is aging and patients are often diagnosed with early-stage lung cancer at an advanced age. Several studies have shown that age is not itself a contraindication for lung cancer surgery, and therefore, more and more octogenarians with early-stage lung cancer are undergoing surgery with curative intent. However, octogenarians present some peculiarities that make surgical treatment more challenging, so an accurate preoperative selection is mandatory. In recent years, new artificial intelligence techniques have spread worldwide in the diagnosis, treatment, and therapy of lung cancer, with increasing clinical applications. However, there is still no evidence coming out from trials specifically designed to assess the potential of artificial intelligence in the preoperative evaluation of octogenarian patients. The aim of this narrative review is to investigate, through the analysis of the available international literature, the advantages and implications that these tools may have in the preoperative assessment of this particular category of frail patients. In fact, these tools could represent an important support in the decision-making process, especially in octogenarian patients in whom the diagnostic and therapeutic options are often questionable. However, these technologies are still developing, and a strict human-led process is mandatory.

show abstract

Section: Radiomics Modelsupporting

confidence: 80%

Section: Spread Through Air Spacesmentioning

confidence: 99%

Section: Future Applications and Limitationsmentioning

confidence: 99%

See 1 more Smart Citation

Lung Cancer Surgery in Octogenarians: Implications and Advantages of Artificial Intelligence in the Preoperative Assessment

Bassi,

Vaz Sousa,

Zacchini

et al. 2024

Healthcare

View full text Add to dashboard Cite

show abstract

“…Moreover, these findings consistently support the notion that STAS is an independent prognostic factor, regardless of the stage of the tumor [ 16 ]. Accurately predicting the presence of STAS in preoperative noninvasive imaging studies can aid preoperative surgical decision-making, and treatment planning and provide a basis for clinical surgery planning [ 17 ]. Previous research has indicated that CT imaging features help predict the STAS phenomenon and determine the appropriate surgical strategy before operation [ 18 ].…”

Section: Discussionmentioning

confidence: 99%

“…A strong correlation exists between nodule size and STAS either. It was concluded that the STAS-positive nodules appeared to have a significantly larger size than the STAS-negative lesions [ 17 ]. According to Margerie et al, subsolid pulmonary adenocarcinomas with histological evidence of STAS were found to be larger than those without [ 22 ].…”

Section: Discussionmentioning

confidence: 99%

Predictive value of CT and 18F-FDG PET/CT features on spread through air space in lung adenocarcinoma

Li,

Liu

et al. 2024

BMC Cancer

View full text Add to dashboard Cite

Background Lung adenocarcinoma, a leading cause of cancer-related mortality, demands precise prognostic indicators for effective management. The presence of spread through air space (STAS) indicates adverse tumor behavior. However, comparative differences between 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography(PET)/computed tomography(CT) and CT in predicting STAS in lung adenocarcinoma remain inadequately explored. This retrospective study analyzes preoperative CT and 18F-FDG PET/CT features to predict STAS, aiming to identify key predictive factors and enhance clinical decision-making. Methods Between February 2022 and April 2023, 100 patients (108 lesions) who underwent surgery for clinical lung adenocarcinoma were enrolled. All these patients underwent 18F-FDG PET/CT, thin-section chest CT scan, and pathological biopsy. Univariate and multivariate logistic regression was used to analyze CT and 18F-FDG PET/CT image characteristics. Receiver operating characteristic curve analysis was performed to identify a cut-off value. Results Sixty lesions were positive for STAS, and 48 lesions were negative for STAS. The STAS-positive was frequently observed in acinar predominant. However, STAS-negative was frequently observed in minimally invasive adenocarcinoma. Univariable analysis results revealed that CT features (including nodule type, maximum tumor diameter, maximum solid component diameter, consolidation tumor ratio, pleural indentation, lobulation, spiculation) and all 18F-FDG PET/CT characteristics were statistically significant difference in STAS-positive and STAS-negative lesions. And multivariate logistic regression results showed that the maximum tumor diameter and SUVmax were the independent influencing factors of CT and 18F-FDG PET/CT in STAS, respectively. The area under the curve of maximum tumor diameter and SUVmax was 0.68 vs. 0.82. The cut-off value for maximum tumor diameter and SUVmax was 2.35 vs. 5.05 with a sensitivity of 50.0% vs. 68.3% and specificity of 81.2% vs. 87.5%, which showed that SUVmax was superior to the maximum tumor diameter. Conclusion The radiological features of SUVmax is the best model for predicting STAS in lung adenocarcinoma. These radiological features could predict STAS with excellent specificity but inferior sensitivity.

show abstract

CT-Based Intratumoral and Peritumoral Radiomics Nomograms for the Preoperative Prediction of Spread Through Air Spaces in Clinical Stage IA Non-small Cell Lung Cancer

Wang,

Lyu,

et al. 2024

J Digit Imaging. Inform. med.

View full text Add to dashboard Cite

The study aims to investigate the value of intratumoral and peritumoral radiomics and clinical-radiological features for predicting spread through air spaces (STAS) in patients with clinical stage IA non-small cell lung cancer (NSCLC). A total of 336 NSCLC patients from our hospital were randomly divided into the training cohort (n = 236) and the internal validation cohort (n = 100) at a ratio of 7:3, and 69 patients from the other two external hospitals were collected as the external validation cohort. Univariate and multivariate analyses were used to select clinical-radiological features and construct a clinical model. The GTV, PTV5, PTV10, PTV15, PTV20, GPTV5, GPTV10, GPTV15, and GPTV20 models were constructed based on intratumoral and peritumoral (5 mm, 10 mm, 15 mm, 20 mm) radiomics features. Additionally, the radscore of the optimal radiomics model and clinical-radiological predictors were used to construct a combined model and plot a nomogram. Lastly, the ROC curve and AUC value were used to evaluate the diagnostic performance of the model. Tumor density type (OR = 6.738) and distal ribbon sign (OR = 5.141) were independent risk factors for the occurrence of STAS. The GPTV10 model outperformed the other radiomics models, and its AUC values were 0.887, 0.876, and 0.868 in the three cohorts. The AUC values of the combined model constructed based on GPTV10 radscore and clinical-radiological predictors were 0.901, 0.875, and 0.878. DeLong test results revealed that the combined model was superior to the clinical model in the three cohorts. The nomogram based on GPTV10 radscore and clinical-radiological features exhibited high predictive efficiency for STAS status in NSCLC.

show abstract

3D convolutional neural network model from contrast-enhanced CT to predict spread through air spaces in non-small cell lung cancer

Cited by 15 publications

References 37 publications

Lung Cancer Surgery in Octogenarians: Implications and Advantages of Artificial Intelligence in the Preoperative Assessment

Lung Cancer Surgery in Octogenarians: Implications and Advantages of Artificial Intelligence in the Preoperative Assessment

Predictive value of CT and 18F-FDG PET/CT features on spread through air space in lung adenocarcinoma

CT-Based Intratumoral and Peritumoral Radiomics Nomograms for the Preoperative Prediction of Spread Through Air Spaces in Clinical Stage IA Non-small Cell Lung Cancer

Contact Info

Product

Resources

About