A neural network-based diagnostic method for solitary pulmonary nodules

Yeh, Chinson; Lin, Chen-Liang; Wu, Ming-Ting; Yen, Chen‐Wen; Wang, Jen-Feng

doi:10.1016/j.neucom.2007.11.009

Cited by 10 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Offline analyses of CT datasets were performed on an institutionally developed program. Segmentation of calcified plaque was based on a semi-automated algorithm called regional growing method[ 26 ], modified and customized by C.W.Y [ 27 ] for efficient assignment of each calcified plaque. The entire coronary arterial tree was inspected for the presence of calcified plaques.…”

Section: Methodsmentioning

confidence: 99%

CT Attenuation Features of Individual Calcified Coronary Plaque: Differences among Asymptomatic, Stable Angina Pectoris, and Acute Coronary Syndrome Groups

Lin

et al. 2015

PLoS ONE

Self Cite

View full text Add to dashboard Cite

BackgroundCoronary artery calcium (CAC) assessed by non-contrast cardiac CT has been shown to be an independent factor from the Framingham risk factors in predicting cardiovascular events. However, many patients with acute coronary syndrome (ACS) have low CAC score. A recent study that re-analyzed the previous CAC CT scan of MESA cohort showed that in subjects with global lower density, CAC was associated with higher risk of ACS. We aimed to further evaluate the characteristics of CAC attenuation features in ACS subjects, in comparison to asymptomatic and stable angina pectoris (SAP) groups.MethodsIn a period of 18 months, 524 consecutive subjects received standard CAC CT scans in our department; 278 of 524 subjects with presence of CAC (225 men, age = 60.6±9.5 years; ACS = 41, SAP = 78, asymptomatic = 159) were enrolled. Agatston score, number of plaques (NP) per subject and mean (HMEAN) and standard deviation (HSD) of attenuation of each calcified plaque were measured. Three regression models to distinguish the groups were built: model 1, conventional risk factors only; model 2, Agatston score plus model 1; model 3, plaque attenuation features plus model 2.ResultsAgatston score in ACS group (median = 112.9) was higher than in the asymptomatic group (median = 54.4, P = 0.028) and similar to the SAP group (median = 237.8, P = 0.428). Calcified plaques in the ACS group showed lower (HMEAN = 180.5) and more homogenous (HSD = 31.2) attenuation than those of the asymptomatic group (HMEAN = 205.9, P = 0.002; HSD = 52.4, P = 0.006) and the SAP group (HMEAN = 204.1, P = 0.016; HSD = 54.4, P = 0.011). Model 3 significantly improved the distinction between ACS and asymptomatic groups (area under curve [AUC] = 0.93) as compared to model 2 (AUC = 0.83, P = 0.003) and model 1 (AUC = 0.79, P = 0.001).ConclusionsCalcified plaques in the ACS group were characteristically of low and homogenous CT attenuation. With validation in a large cohort, analysis of CT attenuation features may improve risk stratification of ACS using CAC CT scan.

show abstract

Section: Methodsmentioning

confidence: 99%

CT Attenuation Features of Individual Calcified Coronary Plaque: Differences among Asymptomatic, Stable Angina Pectoris, and Acute Coronary Syndrome Groups

Lin

et al. 2015

PLoS ONE

Self Cite

View full text Add to dashboard Cite

show abstract

“…Way et al [4] extracted morphological, surface and texture features from 256 lung nodules, and established a linear discriminant analysis. A neural network-based computer-aided diagnosis method of lung nodule diagnosis by combining morphometry and perfusion characteristics to predict characteristics of solitary pulmonary nodules was introduced by Yeh et al [5]. In another study, McCarville et al [6] collected 81 pulmonary nodules, bases on CT findings to differ benign and malignant nature of pulmonary nodules in pediatric patients whereas Wang et al [7] used the gray level co-occurrence matrix and the multi-level model to predict characteristics of pulmonary nodules.…”

Section: Introductionmentioning

confidence: 99%

Computer-Aided Diagnosis for Early-Stage Lung Cancer Based on Longitudinal and Balanced Data

et al. 2013

View full text Add to dashboard Cite

BackgroundLung cancer is one of the most common forms of cancer resulting in over a million deaths per year worldwide. Typically, the problem can be approached by developing more discriminative diagnosis methods. In this paper, computer-aided diagnosis was used to facilitate the prediction of characteristics of solitary pulmonary nodules in CT of lungs to diagnose early-stage lung cancer.MethodsThe synthetic minority over-sampling technique (SMOTE) was used to account for raw data in order to balance the original training data set. Curvelet-transformation textural features, together with 3 patient demographic characteristics, and 9 morphological features were used to establish a support vector machine (SVM) prediction model. Longitudinal data as the test data set was used to evaluate the classification performance of predicting early-stage lung cancer.ResultsUsing the SMOTE as a pre-processing procedure, the original training data was balanced with a ratio of malignant to benign cases of 1∶1. Accuracy based on cross-evaluation for the original unbalanced data and balanced data was 80% and 97%, respectively. Based on Curvelet-transformation textural features and other features, the SVM prediction model had good classification performance for early-stage lung cancer, with an area under the curve of the SVMs of 0.949 (P<0.001). Textural feature (standard deviation) showed benign cases had a higher change in the follow-up period than malignant cases.ConclusionsWith textural features extracted from a Curvelet transformation and other parameters, a sensitive support vector machine prediction model can increase the rate of diagnosis for early-stage lung cancer. This scheme can be used as an auxiliary tool to differentiate between benign and malignant early-stage lung cancers in CT images.

show abstract

“…It is critical to detect and diagnose solitary pulmonary nodules (SPNs), since they are relevant to the choice of treatment and the survival rate [3]. Many computer-aided methods (CAD) have been proposed for SPNs [4][5][6][7][8][9]. In the CAD system, texture is a popular descriptor in the analysis and interpretation of images [10].…”

Section: Introductionmentioning

confidence: 99%

Combination of Radiological and Gray Level Co-occurrence Matrix Textural Features Used to Distinguish Solitary Pulmonary Nodules by Computed Tomography

Sun

Wang

et al. 2013

J Digit Imaging

View full text Add to dashboard Cite

The objective of this study was to investigate the method of the combination of radiological and textural features for the differentiation of malignant from benign solitary pulmonary nodules by computed tomography. Features including 13 gray level co-occurrence matrix textural features and 12 radiological features were extracted from 2,117 CT slices, which came from 202 (116 malignant and 86 benign) patients. Lasso-type regularization to a nonlinear regression model was applied to select predictive features and a BP artificial neural network was used to build the diagnostic model. Eight radiological and two textural features were obtained after the Lasso-type regularization procedure. Twelve radiological features alone could reach an area under the ROC curve (AUC) of 0.84 in differentiating between malignant and benign lesions. The 10 selected characters improved the AUC to 0.91. The evaluation results showed that the method of selecting radiological and textural features appears to yield more effective in the distinction of malignant from benign solitary pulmonary nodules by computed tomography.

show abstract

A neural network-based diagnostic method for solitary pulmonary nodules

Cited by 10 publications

References 36 publications

CT Attenuation Features of Individual Calcified Coronary Plaque: Differences among Asymptomatic, Stable Angina Pectoris, and Acute Coronary Syndrome Groups

CT Attenuation Features of Individual Calcified Coronary Plaque: Differences among Asymptomatic, Stable Angina Pectoris, and Acute Coronary Syndrome Groups

Computer-Aided Diagnosis for Early-Stage Lung Cancer Based on Longitudinal and Balanced Data

Combination of Radiological and Gray Level Co-occurrence Matrix Textural Features Used to Distinguish Solitary Pulmonary Nodules by Computed Tomography

Contact Info

Product

Resources

About