A Cascaded Neural Network for Staging in Non-Small Cell Lung Cancer Using Pre-Treatment CT.
Ontology highlight
ABSTRACT: Background and aim
Tumor staging in non-small cell lung cancer (NSCLC) is important for treatment and prognosis. Staging involves expert interpretation of imaging, which we aim to automate with deep learning (DL). We proposed a cascaded DL method comprised of two steps to classification between early- and advanced-stage NSCLC using pretreatment computed tomography.Methods
We developed and tested a DL model to classify between early- and advanced-stage using training (n = 90), validation (n = 8), and two test (n = 37, n = 26) cohorts obtained from the public domain. The first step adopted an autoencoder network to compress the imaging data into latent variables and the second step used the latent variable to classify the stages using the convolutional neural network (CNN). Other DL and machine learning-based approaches were compared.Results
Our model was tested in two test cohorts of CPTAC and TCGA. In CPTAC, our model achieved accuracy of 0.8649, sensitivity of 0.8000, specificity of 0.9412, and area under the curve (AUC) of 0.8206 compared to other approaches (AUC 0.6824-0.7206) for classifying between early- and advanced-stages. In TCGA, our model achieved accuracy of 0.8077, sensitivity of 0.7692, specificity of 0.8462, and AUC of 0.8343.Conclusion
Our cascaded DL model for classification NSCLC patients into early-stage and advanced-stage showed promising results and could help future NSCLC research.
SUBMITTER: Choi J
PROVIDER: S-EPMC8229025 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PurposeA large chunk of lung cancers are of the type non-small cell lung cancer (NSCLC). Both the treatment planning and patients' prognosis depend greatly on factors like AJCC staging which is an abstraction over TNM staging. Many significant efforts have so far been made towards automated staging of NSCLC, but the groundbreaking application of a deep neural networks (DNNs) is yet to be observed in this domain of study. DNN is capable of achieving higher level of accuracy than the traditional artificial neural networks (ANNs) as it uses deeper layers of convolutional neural network (CNN). The objective of the present study is to propose a simple yet fast CNN model combined with recurrent neural network (RNN) for automated AJCC staging of NSCLC and to compare the outcome with a few standard machine learning algorithms along with a few similar studies.MethodsThe NSCLC radiogenomics collection from the cancer imaging archive (TCIA) dataset was considered for the study. The tumor images were refined and filtered by resizing, enhancing, de-noising, etc. The initial image processing phase was followed by texture based image segmentation. The segmented images were fed into a hybrid feature detection and extraction model which was comprised of two sequential phases: maximally stable extremal regions (MSER) and the speeded up robust features (SURF). After a prolonged experiment, the desired CNN-RNN model was derived and the extracted features were fed into the model.ResultsThe proposed CNN-RNN model almost outperformed the other machine learning algorithms under consideration. The accuracy remained steadily higher than the other contemporary studies.ConclusionThe proposed CNN-RNN model performed commendably during the study. Further studies may be carried out to refine the model and develop an improved auxiliary decision support system for oncologists and radiologists.
| S-EPMC6667593 | biostudies-literature
Project description:ObjectivesTo develop and validate a PET/CT nomogram for preoperative estimation of lymph node (LN) staging in patients with non-small cell lung cancer (NSCLC).MethodsA total of 263 pathologically confirmed LNs from 124 patients with NCSLC were retrospectively analyzed. Positron-emission tomography/computed tomography (PET/CT) examination was performed before treatment according to the clinical schedule. In the training cohort (N = 185), malignancy-related features, such as SUVmax, short-axis diameter (SAD), and CT radiomics features, were extracted from the regions of LN based on the PET/CT scan. The Minimum-Redundancy Maximum-Relevance (mRMR) algorithm and the Least Absolute Shrinkage and Selection Operator (LASSO) regression model were used for feature selection and radiomics score building. The radiomics score (Rad-Score) and SUVmax were incorporated in a PET/CT nomogram using the multivariable logistic regression analysis. The performance of the proposed model was evaluated with discrimination, calibration, and clinical application in an independent testing cohort (N = 78).ResultsThe radiomics scores consisting of 14 selected features were significantly associated with LN status for both training cohort with AUC of 0.849 (95% confidence interval (CI), 0.796-0.903) and testing cohort with AUC of 0.828 (95% CI, 0.782-0.919). The PET/CT nomogram incorporating radiomics score and SUVmax showed moderate improvement of the efficiency with AUC of 0.881 (95% CI, 0.834-0.928) in the training cohort and AUC of 0.872 (95% CI, 0.797-0.946) in the testing cohort. The decision curve analysis indicated that the PET/CT nomogram was clinically useful.ConclusionThe PET/CT nomogram, which incorporates Rad-Score and SUVmax, can improve the diagnostic performance of LN metastasis.Key points• The PET/CT nomogram (Int-Score) based on lymph node (LN) PET/CT images can reliably predict LN status in NSCLC. • Int-Score is a relatively objective diagnostic method, which can play an auxiliary role in the process of clinicians making treatment decisions.
| S-EPMC8270849 | biostudies-literature
Project description:Therapeutic decisions in non-small cell lung cancer (NSCLC) are stage-dependent, and, consequently, changes in an individual's stage carry potential for substantial alterations in management. Malignancy-related disturbances of the circulomic inflammatory environment may affect platelets quantitatively, ultimately leading to changes in tumor characteristics. Our objective was to identify circulomic characteristics associated with upstaging among chemotherapy-naïve patients with resected NSCLC and to assess the consequent impact on overall survival (OS). A retrospective review of a prospectively maintained thoracic surgery database was performed, identifying chemotherapy-naïve patients who underwent resection of clinical stage I-III NSCLC between 1998 and 2021. Clinicopathologic characteristics were gathered; circulomic variables comprised of platelet and lymphocyte count from the last blood draw prior to resection. Platelet-to-lymphocyte ratio (PLR) was calculated. A multivariate model evaluated variables that might affect upstaging. Kaplan-Meier analysis was performed to assess OS. A total of 4,141 patients met inclusion criteria (median age: 67.0 years) among whom the sex distribution was fairly equal (2,189 female, 52.9%), and 1,016 (24.5%) individuals were upstaged. Patients with elevated PLR were found to have reduced risk of upstaging [odds ratio (OR): 0.757, 95% confidence interval (CI): 0.650-0.882]. Analyses revealed that median OS for patients who were upstaged was 80.0 months compared to 130.7 months among those who weren't upstaged (P<0.0001). PLR appears to predict upstaging in treatment-naïve patients with resected NSCLC. In addition to clinicopathologic characteristics, circulomic variables may provide insight relating to pathologic staging prior to resection. These findings may guide patient counseling regarding survival probability, as well as referral patterns for adjuvant therapy.
| S-EPMC10636430 | biostudies-literature
Project description:BackgroundComputed tomography (CT) is important in the diagnosing of lung cancer. The combination of CT features and artificial intelligence algorithm have been used in the diagnosis of various lung diseases. However, limited studies focused on the relationship between the combination of CT features and artificial intelligence algorithm and lymph node metastasis in non-small cell lung cancer (NSCLC). This study developed an algorithm for lung cancer CT image segmentation based on an artificial neural network model and investigated the role of a nomogram model based on CT images for predicting lymph node metastasis in lung cancer.MethodsWiener filtering and fuzzy enhancement were first used to suppress image noise and improve image contrast. Next, texture features and fractal features were extracted. In the third step, the artificial neural network model was trained and tested according to the best parameters of the network.ResultsThe area under the curve (AUC) of the constructed nomogram model on the training set and the test set were 0.859 (sensitivity, 0.810; specificity, 0.773) and 0.864 (sensitivity, 0.820; specificity, 0.753), respectively. The decision curve indicated that the model had good clinical application value. The lung cancer CT images contained 13 significant regional features of cancer. The best classification function obtained from training and testing data was Levenberg-Marquardt backpropagation. The sensitivity, specificity, and accuracy in the training stage could reach 98.4%, 100%, and 98.6%, respectively, and the corresponding indexes in the test stage reached 90.9%, 100%, and 95.1%, respectively.ConclusionsThe image segmentation algorithm based on the artificial neural network model could extract CT lung cancer lesions efficiently and quasi-determinately, which could be used as an effective tool for radiologists to diagnose lung cancer. The nomogram model based on CT image features and related clinical indicators was an effective method for noninvasive prediction of lymph node metastasis in lung cancer.
| S-EPMC9745522 | biostudies-literature
Project description:PurposeTo evaluate the effect of adding lymph node size to three previously explored artificial neural network (ANN) input parameters (primary tumor maximum standardized uptake value or tumor uptake, tumor size, and nodal uptake at N1, N2, and N3 stations) in the structure of the ANN. The goal was to allow the resulting ANN structure to relate lymph node uptake for size to primary tumor uptake for size in the determination of the status of nodes as human readers do.Materials and methodsThis prospective study was approved by the institutional review board, and informed consent was obtained from all participants. The authors developed a back-propagation ANN with one hidden layer and eight processing units. The data set used to train the network included node and tumor size and uptake from 133 patients with non-small cell lung cancer with surgically proved N status. Statistical analysis was performed with the paired t test.ResultsThe ANN correctly predicted the N stage in 99.2% of cases, compared with 72.4% for the expert reader (P < .001). In categorization of N0 and N1 versus N2 and N3 disease, the ANN performed with 99.2% accuracy versus 92.2% for the expert reader (P < .001).ConclusionThe ANN is 99.2% accurate in predicting surgical-pathologic nodal status with use of four fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT)-derived parameters. Malignant and benign inflammatory lymph nodes have overlapping appearances at FDG PET/CT but can be differentiated by ANNs when the crucial input of node size is used.
| S-EPMC4228715 | biostudies-literature
Project description:BackgroundClinical practice guidelines and re-imbursement schedules vary in the recommended timing of FDG-PET/CT in the diagnostic evaluation of suspected or confirmed lung cancer. The aim was to estimate the probability of requiring more than one invasive test to complete diagnosis and staging in non-small cell lung cancer if FDG-PET/CT was used prior to initial biopsy (FDG-PET/CT First) compared to current Australian funding criteria (CT First).MethodsSingle-centre retrospective study of individuals with pathologically confirmed NSCLC without evidence of metastatic disease on baseline computed tomography (CT) of the chest. Decision tree analysis based on diagnosis and staging approaches estimated the probability of requiring more than one invasive biopsy. A Monte Carlo analysis with 1000 simulations was used to estimate decision tree precision.ResultsAfter exclusions, 115 patients were included with median (IQR) age of 71 (63-79) and 55.6% were male. The majority of cases were early stage (Stage I 43.5%, Stage II 19.1%) and adenocarcinoma (65.2%) histological subtype. The estimated probability of requiring more than one invasive biopsy with FDG-PET/CT prior was 0.12 compared to 0.19 when using the base case CT First scenario. Using the Monte Carlo analysis, the mean (95% CI) probability using the FDG-PET First approach was 0.15 (95%CI 0.12-0.20) versus 0.20 (95% CI 0.15-0.27) for the CT First approach. Only 7.8% had CT Chest-occult metastatic disease on FDG-PET that was accessible by percutaneous biopsy.ConclusionFDG-PET/CT performed prior to initial biopsy may reduce the proportion of people with NSCLC who require more than one biopsy attempt, but the clinical significance and overall cost-utility requires evaluation.
| S-EPMC8252249 | biostudies-literature
Project description:BackgroundThe current preferred approach for surgical mediastinal staging of non-small-cell lung carcinoma is video-assisted mediastinoscopy. An alternative technique in which lymph nodes are resected instead of biopsied is video-assisted mediastinoscopic lymphadenectomy (VAMLA) that is suggested to be superior in detecting N2 disease. Yet, evidence is conflicting and furthermore limited by sample size. The objective was to compare mediastinal staging through VAMLA and video-assisted mediastinoscopy.MethodsA single-center cohort study was conducted. All consecutive patients that underwent surgical mediastinal staging of non-small-cell lung carcinoma by VAMLA (2011 to 2018) were compared to historic video-assisted mediastinoscopy controls (2007 to 2011). Patients with negative surgical mediastinal staging underwent subsequent anatomical resection with systematic regional lymphadenectomy. Primary outcome was the sensitivity and negative predictive value for detecting N2 disease.ResultsTwo-hundred-sixty-nine video-assisted mediastinoscopic lymphadenectomies and 118 video-assisted mediastinoscopies were performed. The prevalence of N2 disease was 20% and 26% respectively in the VAMLA and video-assisted mediastinoscopy group, while the rate of unforeseen pN2 resulting from lymph node dissection during anatomical resection was 4% and 11%, respectively. Invasive staging using VAMLA demonstrated superior sensitivity of 0.82 and a negative predictive value of 0.96 when compared to video-assisted mediastinoscopy (0.62 and 0.89, respectively), offering a 64% decrease in risk of unforeseen pN2 following anatomical resection. However, VAMLA is also associated with a 75% risk increase on complications (P=0.36).ConclusionsWe conclude that performing invasive mediastinal lymph node assessment for staging of non-small-cell lung carcinoma, VAMLA should be the preferred technique with superior sensitivity and negative predictive value in detecting N2 disease. Though, VAMLA is also associated with an increased risk of complications.
| S-EPMC8435384 | biostudies-literature
Project description:The image texture features obtained from 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) images of non-small cell lung cancer (NSCLC) have revealed tumor heterogeneity. A combination of genomic data and radiomics may improve the prediction of tumor prognosis. This study aimed to predict NSCLC metastasis using a graph neural network (GNN) obtained by combining a protein-protein interaction (PPI) network based on gene expression data and image texture features. 18F-FDG PET/CT images and RNA sequencing data of 93 patients with NSCLC were acquired from The Cancer Imaging Archive. Image texture features were extracted from 18F-FDG PET/CT images and area under the curve receiver operating characteristic curve (AUC) of each image feature was calculated. Weighted gene co-expression network analysis (WGCNA) was used to construct gene modules, followed by functional enrichment analysis and identification of differentially expressed genes. The PPI of each gene module and genes belonging to metastasis-related processes were converted via a graph attention network. Images and genomic features were concatenated. The GNN model using PPI modules from WGCNA and metastasis-related functions combined with image texture features was evaluated quantitatively. Fifty-five image texture features were extracted from 18F-FDG PET/CT, and radiomic features were selected based on AUC (n = 10). Eighty-six gene modules were clustered by WGCNA. Genes (n = 19) enriched in the metastasis-related pathways were filtered using DEG analysis. The accuracy of the PPI network, derived from WGCNA modules and metastasis-related genes, improved from 0.4795 to 0.5830 (p < 2.75 × 10-12). Integrating PPI of four metastasis-related genes with 18F-FDG PET/CT image features in a GNN model elevated its accuracy over a without image feature model to 0.8545 (95% CI = 0.8401-0.8689, p-value < 0.02). This model demonstrated significant enhancement compared to the model using PPI and 18F-FDG PET/CT derived from WGCNA (p-value < 0.02), underscoring the critical role of metastasis-related genes in prediction model. The enhanced predictive capability of the lymph node metastasis prediction GNN model for NSCLC, achieved through the integration of comprehensive image features with genomic data, demonstrates promise for clinical implementation.
| S-EPMC10815846 | biostudies-literature
Project description:The evolution of next-generation sequencing technology has resulted in a generation of large amounts of cancer genomic data. Therefore, increasingly complex techniques are required to appropriately analyze this data in order to determine its clinical relevance. In this study, we applied a neural network-based technique to analyze data from The Cancer Genome Atlas and extract useful microRNA (miRNA) features for predicting the prognosis of patients with lung adenocarcinomas (LUAD). Using the Cascaded Wx platform, we identified and ranked miRNAs that affected LUAD patient survival and selected the two top-ranked miRNAs (miR-374a and miR-374b) for measurement of their expression levels in patient tumor tissues and in lung cancer cells exhibiting an altered epithelial-to-mesenchymal transition (EMT) status. Analysis of miRNA expression from tumor samples revealed that high miR-374a/b expression was associated with poor patient survival rates. In lung cancer cells, the EMT signal induced miR-374a/b expression, which, in turn, promoted EMT and invasiveness. These findings demonstrated that this approach enabled effective identification and validation of prognostic miRNA markers in LUAD, suggesting its potential efficacy for clinical use.
| S-EPMC7409139 | biostudies-literature
Project description:PurposeOur goal was to develop a nomogram by exploiting intratumour heterogeneity on CT and PET images from routine (18)F-FDG PET/CT acquisitions to identify patients with the poorest prognosis.MethodsThis retrospective study included 116 patients with NSCLC stage I, II or III and with staging (18)F-FDG PET/CT imaging. Primary tumour volumes were delineated using the FLAB algorithm and 3D Slicer™ on PET and CT images, respectively. PET and CT heterogeneities were quantified using texture analysis. The reproducibility of the CT features was assessed on a separate test-retest dataset. The stratification power of the PET/CT features was evaluated using the Kaplan-Meier method and the log-rank test. The best standard metric (functional volume) was combined with the least redundant and most prognostic PET/CT heterogeneity features to build the nomogram.ResultsPET entropy and CT zone percentage had the highest complementary values with clinical stage and functional volume. The nomogram improved stratification amongst patients with stage II and III disease, allowing identification of patients with the poorest prognosis (clinical stage III, large tumour volume, high PET heterogeneity and low CT heterogeneity).ConclusionIntratumour heterogeneity quantified using textural features on both CT and PET images from routine staging (18)F-FDG PET/CT acquisitions can be used to create a nomogram with higher stratification power than staging alone.
| S-EPMC5409954 | biostudies-literature