scholarly journals Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging

2019 ◽  
Author(s):  
Xue Sha ◽  
Guan Zhong Gong ◽  
Qing Tao Qiu ◽  
Jing Hao Duan ◽  
Deng Wang Li ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Predictive Value ◽  
Multivariate Logistic Regression Analysis ◽  
Ct Images ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase ◽  
Nsclc Patients

Abstract Background: We aimed to develop radiomic models based on different phases of computed tomography (CT) imaging and to investigate the efficacy of models for diagnosing mediastinal metastatic lymph nodes (LNs) in non-small cell lung cancer (NSCLC). Methods: Eighty-six NSCLC patients were enrolled in this study, and we selected 231 mediastinal LNs confirmed by pathology results as the subjects which were divided into training (n=163) and validation cohorts (n=68). The regions of interest (ROIs) were delineated on CT scans in the plain phase, arterial phase and venous phase, respectively. Radiomic features were extracted from the CT images in each phase. A least absolute shrinkage and selection operator (LASSO) algorithm was used to select features, and multivariate logistic regression analysis was used to build models. We constructed six models (orders 1-6) based on the radiomic features of the single- and dual-phase CT images. The performance of the radiomic model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV). Results: A total of 846 features were extracted from each ROI, and 10, 9, 5, 2, 2, and 9 features were chosen to develop models 1-6, respectively. All of the models showed excellent discrimination, with AUCs greater than 0.8. The plain CT radiomic model, model 1, yielded the highest AUC, specificity, accuracy and PPV, which were 0.926 and 0.925; 0.860 and 0.769; 0.871 and 0.882; and 0.906 and 0.870 in the training and validation sets, respectively. When the plain and venous phase CT radiomic features were combined with the arterial phase CT images, the sensitivity increased from 0.879 and 0.919 to 0.949 and 0979 and the NPV increased from 0.821 and 0.789 to 0.878 and 0.900 in the training group, respectively. Conclusions: All of the CT radiomic models based on different phases all showed high accuracy and precision for the diagnosis of LN metastasis (LNM) in NSCLC patients. When combined with arterial phase CT, the sensitivity and NPV of the model was be further improved.

scholarly journals Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging

2020 ◽  
Author(s):  
Xue Sha ◽  
Guan Zhong Gong ◽  
Qing Tao Qiu ◽  
Jing Hao Duan ◽  
Deng Wang Li ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Predictive Value ◽  
Multivariate Logistic Regression Analysis ◽  
Ct Images ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase ◽  
Nsclc Patients

Abstract Background: We aimed to develop radiomic models based on different phases of computed tomography (CT) imaging and to investigate the efficacy of models for diagnosing mediastinal metastatic lymph nodes (LNs) in non-small cell lung cancer (NSCLC). Methods: Eighty-six NSCLC patients were enrolled in this study, and we selected 231 mediastinal LNs confirmed by pathology results as the subjects which were divided into training (n=163) and validation cohorts (n=68). The regions of interest (ROIs) were delineated on CT scans in the plain phase, arterial phase and venous phase, respectively. Radiomic features were extracted from the CT images in each phase. A least absolute shrinkage and selection operator (LASSO) algorithm was used to select features, and multivariate logistic regression analysis was used to build models. We constructed six models (orders 1-6) based on the radiomic features of the single- and dual-phase CT images. The performance of the radiomic model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV). Results: A total of 846 features were extracted from each ROI, and 10, 9, 5, 2, 2, and 9 features were chosen to develop models 1-6, respectively. All of the models showed excellent discrimination, with AUCs greater than 0.8. The plain CT radiomic model, model 1, yielded the highest AUC, specificity, accuracy and PPV, which were 0.926 and 0.925; 0.860 and 0.769; 0.871 and 0.882; and 0.906 and 0.870 in the training and validation sets, respectively. When the plain and venous phase CT radiomic features were combined with the arterial phase CT images, the sensitivity increased from 0.879 and 0.919 to 0.949 and 0979 and the NPV increased from 0.821 and 0.789 to 0.878 and 0.900 in the training group, respectively. Conclusions: All of the CT radiomic models based on different phases all showed high accuracy and precision for the diagnosis of LN metastasis (LNM) in NSCLC patients. When combined with arterial phase CT, the sensitivity and NPV of the model was be further improved.

scholarly journals Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging

2020 ◽  
Author(s):  
Xue Sha ◽  
Guan Zhong Gong ◽  
Qing Tao Qiu ◽  
Jing Hao Duan ◽  
Deng Wang Li ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Predictive Value ◽  
Multivariate Logistic Regression Analysis ◽  
Ct Images ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase ◽  
Nsclc Patients

Abstract Background: We aimed to develop radiomic models based on different phases of computed tomography (CT) imaging and to investigate the efficacy of models for diagnosing mediastinal metastatic lymph nodes (LNs) in non-small cell lung cancer (NSCLC). Methods: Eighty-six NSCLC patients were enrolled in this study, and we selected 231 mediastinal LNs confirmed by pathology results as the subjects which were divided into training (n=163) and validation cohorts (n=68). The regions of interest (ROIs) were delineated on CT scans in the plain phase, arterial phase and venous phase, respectively. Radiomic features were extracted from the CT images in each phase. A least absolute shrinkage and selection operator (LASSO) algorithm was used to select features, and multivariate logistic regression analysis was used to build models. We constructed six models (orders 1-6) based on the radiomic features of the single- and dual-phase CT images. The performance of the radiomic model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV). Results: A total of 846 features were extracted from each ROI, and 10, 9, 5, 2, 2, and 9 features were chosen to develop models 1-6, respectively. All of the models showed excellent discrimination, with AUCs greater than 0.8. The plain CT radiomic model, model 1, yielded the highest AUC, specificity, accuracy and PPV, which were 0.926 and 0.925; 0.860 and 0.769; 0.871 and 0.882; and 0.906 and 0.870 in the training and validation sets, respectively. When the plain and venous phase CT radiomic features were combined with the arterial phase CT images, the sensitivity increased from 0.879 and 0.919 to 0.949 and 0979 and the NPV increased from 0.821 and 0.789 to 0.878 and 0.900 in the training group, respectively. Conclusions: All of the CT radiomic models based on different phases all showed high accuracy and precision for the diagnosis of LN metastasis (LNM) in NSCLC patients. When combined with arterial phase CT, the sensitivity and NPV of the model was be further improved.

scholarly journals Discrimination of mediastinal metastatic lymph nodes in NSCLC based on radiomic features in different phases of CT imaging

2019 ◽  
Author(s):  
Xue Sha ◽  
Guan Zhong Gong ◽  
Qing Tao Qiu ◽  
Jing Hao Duan ◽  
Deng Wang Li ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Predictive Value ◽  
Characteristic Curve ◽  
Multivariate Logistic Regression Analysis ◽  
Ct Images ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase

Abstract Background: We aimed to develop radiomic models based on different phases of computed tomography (CT) imaging and to investigate the efficacy of models for diagnosing mediastinal metastatic lymph nodes (LNs) in non-small cell lung cancer (NSCLC). Methods: We selected 231 mediastinal LNs confirmed by pathology results as the subjects, which were divided into training (n=163) and validation cohorts (n=68). The regions of interest (ROIs) were delineated on CT scans in the plain phase, arterial phase and venous phase, respectively. Radiomic features were extracted from the CT images in each phase. A least absolute shrinkage and selection operator (LASSO) algorithm was used to select features, and multivariate logistic regression analysis was used to build models. We constructed six models (orders 1-6) based on the radiomic features of the single- and dual-phase CT images. The performance of the radiomic model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV). Results: A total of 846 features were extracted from each ROI, and 10, 9, 5, 2, 2, and 9 features were chosen to develop models 1-6, respectively. All of the models showed excellent discrimination, with AUCs greater than 0.8. The plain CT radiomic model, model 1, yielded the highest AUC, specificity, accuracy and PPV, which were 0.926 and 0.925; 0.860 and 0.769; 0.871 and 0.882; and 0.906 and 0.870 in the training and validation sets, respectively. When the plain and venous phase CT radiomic features were combined with the arterial phase CT images, the sensitivity increased from 0.879 and 0.919 to 0.949 and 0979 and the NPV increased from 0.821 and 0.789 to 0.878 and 0.900 in the training group, respectively. Conclusions: All of the CT radiomic models based on different phases all showed high accuracy and precision for the diagnosis of LN metastasis (LNM) in NSCLC patients. When combined with arterial phase CT, the sensitivity and NPV of the model was be further improved.

scholarly journals Differentiation of mediastinal metastatic lymph nodes in NSCLC based on radiomic features of different phase CT imaging

2019 ◽  
Author(s):  
Xue Sha ◽  
Guan Zhong Gong ◽  
Qing Tao Qiu ◽  
Jing Hao Duan ◽  
Deng Wang Li ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Predictive Value ◽  
Characteristic Curve ◽  
Multivariate Logistic Regression Analysis ◽  
Ct Images ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase

Abstract Background To develop radiomic models based on different phases of computed tomography (CT) imaging and investigate the efficacy of models to diagnose mediastinal metastatic lymph nodes in non-small cell lung cancer (NSCLC).Methods We selected 231 mediastinal lymph nodes confirmed by pathology results as the subjects, which were divided into training (n=163) and validation cohorts (n=68). The regions of interest (ROIs) were delineated on CT scans of the plain phase, arterial phase and venous phase, respectively. Radiomic features were extracted from the CT images of each phase. Least absolute shrinkage and selection operator (LASSO) was used to select features, and multivariate logistic regression analysis was used to build models. We constructed six models (orders of 1-6) based on radiomic features of the single- and dual-phase CT images. The performance of the radiomic model was evaluated by the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV).Results A total of 846 features were extracted from each ROI, and 10, 9, 5, 2, 2, and 9 features were chosen to develop models 1-6. All of the models showed superior differentiation, with AUCs greater than 0.8. The plain CT radiomic model, model 1, yielded the highest AUC, specificity, accuracy and PPV, which were 0.926 VS 0.925, 0.860 VS 0.769, 0.871 VS 0.882 and 0.906 VS 0.870 in the training and validation sets, respectively. When the plain and venous phase CT radiomic features were combined with the arterial phase CT images, the sensitivity increased from 0.879, 0.919 to 0.949, 0979 and the NPV increased from 0.821, 0.789 to 0.878, 0.900 in the training group, respectively.Conclusion CT radiomic models based on different phases all showed high accuracy and precision in the diagnosis of LNM in NSCLC patients. When combined with arterial phase CT, the sensitivity and NPV of the model can be further improved.

scholarly journals Evaluation of spectral CT in differentiating metastatic from non-metastatic lymph nodes in non-small cell lung cancer(NSCLC)

2020 ◽  
Author(s):  
Funan Wang ◽  
Yanwei Wang ◽  
Gang Guo ◽  
Liuhong Zhu
Keyword(s):  
Lung Cancer ◽  
Lymph Nodes ◽  
Iodine Concentration ◽  
Arterial Phase ◽  
Spectral Ct ◽  
Small Cell Lung ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Venous Phase ◽  
The Difference

Abstract Background Lung cancer is the main cause of tumor-correlated deaths, analysis of lymph nodes is crucial to staging of lung cancer. The purpose of the study is to explore the importance of spectral CT imaging in the difference prognostic of metastatic and non-metastatic mediastinal lymph nodes in non-small cell lung cancer. Methods A retrospective examination of 76 patients with non-small cell lung tumor who underwent spectral CT was performed. Quantifiable GSI (Gemstone spectral imaging) parameters (eg, 40 keV, iodine concentration, water concentration) were calculated in non-contrast, arterial and venous phase in 110 mediastinal lymph nodes using AW4.6 (GE HEALTHCARE, USA). Results The CT values ​​of 40 kev, λHU (The slope of Hounsfield unit curve) and IC(values of iodine concentration values), WC(values of water concentration) measured at the arterial or venous phase were not significantly different from those of metastatic growth lymph nodes (P > 0.05). The net value of Arterial phase (nIAP,net value of iodine concentration in Arterial phase) and vein phase (nIVP, net value of iodine concentration in vein phase) were calculated. The value of nIAP was the difference between IAP (iodine concentration of arterial phase) and INCP (iondine concentration of non-contrast phase), while the value of nIVP was the difference between IVP (iodine concentration of venous phase) and INCP. There stood no noteworthy difference in nIAP amid metastatic lymph nodes and non-metastatic lymph nodes (P = 0.110). There was a substantial difference in nIVP amid metastatic lymph nodes and non-metastatic lymph nodes (P = 0.001). Conclusions Compared with qualitative assessment with conventional CT imaging features, quantitative GSI parameters (nIVP) showed higher accuracy for the preoperative diagnosis of mediastal lymph nodal metastases in patients with NSCLC.

Application of spectral CT imaging in evaluating lymph node metastasis in patients with gastric cancers: initial findings

2018 ◽  
Vol 60 (4) ◽  
pp. 415-424 ◽  
Author(s):  
Zhuping Zhou ◽  
Yu Liu ◽  
Kui Meng ◽  
Wenxian Guan ◽  
Jian He ◽  
...  
Keyword(s):  
Lymph Node ◽  
Lymph Node Metastasis ◽  
Lymph Nodes ◽  
Ct Imaging ◽  
Arterial Phase ◽  
Spectral Ct ◽  
Metastatic Lymph Nodes ◽  
Gastric Cancers ◽  
Node Metastasis ◽  
Metastatic Lymph

Background Traditional computed tomography (CT) can predict the lymph node metastasis of gastric cancers with moderate accuracy; however, investigation of spectral CT imaging in this field is still limited. Purpose To explore the application of spectral CT imaging in evaluating lymph node metastasis in patients with gastric cancers. Material and Methods Twenty-four patients with gastric cancers prospectively underwent spectral CT imaging in the arterial phase. The short and long diameters, material concentrations, and CT values were measured and compared between lymph nodes with and without metastasis. The diagnostic performance of the CT index in identifying metastatic lymph nodes was analyzed with receiver operating characteristic (ROC) analysis. Results A total of 102 lymph nodes (77 metastatic, 25 non-metastatic) were detected on spectral CT imaging with the reference of postoperative pathologic exanimation. The short and long diameters, water/fat concentrations, CT value, and ratio between lymph nodes vs. tumors of metastatic lymph nodes were significantly higher than those of non-metastatic ones (all P < 0.05). With a cut-off of 0.785, the CT ratio of lymph node/tumor on 70-keV monochromatic images yielded an accuracy of 81.4% in differentiating lymph nodes with and without metastasis. Conclusion Spectral CT imaging detects lymph nodes more clearly, and the CT ratio of lymph node/tumor on 70-keV monochromatic images holds great potential in differentiating lymph nodes with and without metastasis, which is more accurate than size measurement.

scholarly journals Usefulness of Hybrid PET/MRI in Clinical Evaluation of Head and Neck Cancer Patients

Cancers ◽  
2020 ◽  
Vol 12 (2) ◽  
pp. 511 ◽  
Author(s):  
Natalia Samolyk-Kogaczewska ◽  
Ewa Sierko ◽  
Dorota Dziemianczyk-Pakiela ◽  
Klaudia Beata Nowaszewska ◽  
Malgorzata Lukasik ◽  
...  
Keyword(s):  
Head And Neck Cancer ◽  
Lymph Nodes ◽  
Head And Neck ◽  
Neck Cancer ◽  
Predictive Value ◽  
Tumor Volume ◽  
Preoperative Staging ◽  
Tumor Diameter ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph

(1) Background: The novel hybrid of positron emission tomography/magnetic resonance (PET/MR) examination has been introduced to clinical practice. The aim of our study was to evaluate PET/MR usefulness in preoperative staging of head and neck cancer (HNC) patients (pts); (2) Methods: Thirty eight pts underwent both computed tomography (CT) and PET/MR examination, of whom 21 pts underwent surgical treatment as first-line therapy and were further included in the present study. Postsurgical tissue material was subjected to routine histopathological (HP) examination with additional evaluation of p16, human papillomavirus (HPV), Epstein-Barr virus (EBV) and Ki67 status. Agreement of clinical and pathological T staging, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of CT and PET/MR in metastatic lymph nodes detection were defined. The verification of dependences between standardized uptake value (SUV value), tumor geometrical parameters, number of metastatic lymph nodes in PET/MR and CT, biochemical parameters, Ki67 index, p16, HPV and EBV status was made with statistical analysis of obtained results; (3) Results: PET/MR is characterized by better agreement in T staging, higher specificity, sensitivity, PPV and NPV of lymph nodes evaluation than CT imaging. Significant correlations were observed between SUVmax and maximal tumor diameter from PET/MR, between SUVmean and CT tumor volume, PET/MR tumor volume, maximal tumor diameter assessed in PET/MR. Other correlations were weak and insignificant; (4) Conclusions: Hybrid PET/MR imaging is useful in preoperative staging of HNC. Further studies are needed.

scholarly journals Comparison of [68Ga]Ga-DOTA-FAPI-04 PET/CT with [18F]FDG PET/CT for Nodule Characterization and Staging in NSCLC Patients

2021 ◽  
Author(s):  
Xin Zhou ◽  
Shuailiang Wang ◽  
Xiaoxia Xu ◽  
Xiangxi Meng ◽  
Huiyuan Zhang ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Fdg Pet ◽  
Lesion Detection ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Imaging Characteristics ◽  
Pet Ct ◽  
Fdg Pet Ct ◽  
Nsclc Patients ◽  
N Staging

Abstract Purpose:The aim of this study is to explore the nodule characterization and staging efficacy of [68Ga]Ga-DOTA-FAPI-04 PET/CT in non-small cell lung cancer (NSCLC) patients and to compare with that of [18F]FDG PET/CT lesion-by-lesion.Methods:Sixty-five patients with clinically diagnosed or suspected NSCLC were enrolled in this prospective study (ChiCTR2000038080). All patients received both [18F]FDG and [68Ga]Ga-DOTA-FAPI-04 PET/CT, and they were assigned into three groups by different research directions as nodule characterization, node (N) staging and metastatic (M) staging. Imaging characteristics in PET/CT of lung nodules and suspected metastatic lesions were obtained and analyzed.Results:In the nodule characterization group, [18F]FDG and [68Ga]Ga-DOTA-FAPI-04 SUVmax ≥ 2.5 was set as the predictor of NSCLC, and the diagnostic sensitivity of [68Ga]Ga-DOTA-FAPI-04 was higher than [18F]FDG (0.88 vs. 0.67). And for adenocarcinoma with partial-solid density, SUVmax of 68Ga-DOTA-FAPI-04 was higher than [18F]FDG with significant differences (4.8 ± 2.8 vs. 2.1 ± 1.1). In N staging group, lymph nodes SUVmax of [68Ga]Ga-DOTA-FAPI-04 was lower than [18F]FDG in nonmetastatic group (3.1 ± 1.3 vs. 6.1 ± 2.3) and higher than [18F]FDG (10.7 ± 4.7 vs. 6.5 ± 3.3) in metastatic group. Set 6 and 1.1 as the cut-off value for [68Ga]Ga-DOTA-FAPI-04 SUVmax and [68Ga]Ga-DOTA-FAPI-04 SUVmax/FDG SUVmax, diagnostic accuracy of metastatic lymph nodes using each criterion and their combination was 95%, 93% and 97% respectively. In multi-metastatic NSCLC patients, [68Ga]Ga-DOTA-FAPI-04 identified more lesions than [18F]FDG (206 vs. 106 lesions) and the uptake value of [68Ga]Ga-DOTA-FAPI-04 was higher too, but no patients’ staging was changed.Conclusion:Compared with [18F]FDG, [68Ga]Ga-DOTA-FAPI-04 PET/CT imaging has higher sensitivity in primary and metastatic lesion detection of NSCLC patients, it also increases the specificity of metastatic lymph nodes diagnosis.

scholarly journals Predictive Value of Lymph Node Ratio in Patients with Pancreatic Cancer Treated with Pancreatoduodenectomy

2021 ◽  
Vol 14 (3) ◽  
Author(s):  
Mahdi Attarian ◽  
Mohsen Aliakbarian ◽  
Ali Taghizadeh kermani ◽  
Fahimeh Attarian ◽  
Sakineh Amouian ◽  
...  
Keyword(s):  
Pancreatic Cancer ◽  
Lymph Node ◽  
Lymph Nodes ◽  
Predictive Value ◽  
Lymph Node Ratio ◽  
Survival Rates ◽  
Whipple Procedure ◽  
Metastatic Lymph Nodes ◽  
Metastatic Lymph ◽  
Node Ratio

Background: Recently, the predictive value of lymph node ratio (LNR, the ratio of metastatic lymph nodes to total examined lymph nodes) has been evaluated in patients with gastrointestinal malignancies, including pancreatic cancer. However, there is not enough evidence about the prognostic value of this factor. Objectives: We aimed at determining the value of LNR in predicting the survival of patients who have undergone the Whipple procedure. Methods: This cohort study was performed on 96 patients with pancreatic cancer undergoing the Whipple procedure during 2014 - 2019. Demographic, clinical, and pathological data of the patients were extracted from their records and patients' survival status was determined through follow-up. LNR and its effect on survival was calculated using the Cox model. Results: Of the 96 eligible patients, 51 (53.13%) were men. The mean age of the patients was 57.1 ± 14.1 (range: 19 - 82) years. The median total lymph nodes examined was 7 (range: 1 - 27), and no metastatic lymph nodes were found in 57 (59.37%) patients. The median involved lymph nodes and LNR were 2 and 0.17, respectively. Patients with LNR > 0.20 had the lowest 1 and 3-year survival rates. Age (P = 0.04), surgical radial margin (P = 0.001), lymph node status (N0, N1) (P = 0.01), and LNR (P = 0.01) were the most important prognostic factors for survival. Conclusions: LNR is a valuable indicator that can be used in patients with lymph node involvement as a prognostic factor for poor survival after the Whipple procedure. The lowest 1, 3, and 5-year survival rates were seen in patients with LNR > 0.20.

Sign in / Sign up

Export Citation Format

Share Document