Exploratory Study of a CT Radiomics Model for the Classification of Small Cell Lung Cancer and Non-small-Cell Lung Cancer

Abstract

Background: Radiomics can quantify tumor phenotypic characteristics non-invasively by applying feature algorithms to medical imaging data. In this study, we investigated the association between radiomics features and the tumor histological subtypes, and we aimed to establish a nomogram for the classification of small cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC). Methods: This was a retrospective single center study. In total, 468 cases including 202 patients with SCLC and 266 patients with NSCLC were enrolled in our study, and were randomly divided into a training set (n = 327) and a validation set (n = 141) in a 7:3 ratio. The clinical data of the patients, including age, sex, smoking history, tumor maximum diameter, clinical stage, and serum tumor markers, were collected. All patients underwent enhanced computed tomography (CT) scans, and all lesions were pathologically confirmed. A radiomics signature was generated from the training set using the least absolute shrinkage and selection operator algorithm. Independent risk factors were identified by multivariate logistic regression analysis, and a radiomics nomogram based on the radiomics signature and clinical features was constructed. The capability of the nomogram was evaluated in the training set and validated in the validation set. Results: Fourteen of 396 radiomics parameters were screened as important factors for establishing the radiomics model. The radiomics signature performed well in differentiating SCLC and NSCLC, with an area under the curve (AUC) of 0.86 (95% CI: 0.82-0.90) in the training set and 0.82 (95% CI: 0.75-0.89) in the validation set. The radiomics nomogram had better predictive performance [AUC = 0.94 (95% CI: 0.90-0.98) in the validation set] than the clinical model [AUC = 0.86 (95% CI: 0.80-0.93)] and the radiomics signature [AUC = 0.82 (95% CI: 0.75-0.89)], and the accuracy was 86.2% (95% CI: 0.79-0.92) in the validation set. Conclusion: The enhanced CT radiomics signature performed well in the classification of SCLC and NSCLC. The nomogram based on the radiomics signature and clinical factors has better diagnostic performance for the classification of SCLC and NSCLC than the simple application of the radiomics signature.

Keywords: X ray; classification; lung cancer histology; quantitative imaging; radiomics; tomography.

Figure 1

Flowchart of the study group inclusion process.

Figure 2

Flow chart of radiomics implementation in this study.

Figure 3

(A) The binomial deviation from the lasso regression cross-validation model is plotted as a log (λ) function by using the 10-fold cross-validation method. The y-axis represents binomial deviation, the lower x-axis represents log (λ), and the numbers above the x-axis represent the average number of predictive variables. The red dot represents the average deviation value of each model with a given λ, while the vertical bar of the red dot represents the upper and lower limit values of the deviation. The vertical dotted line represents the log (λ) value corresponding to the best λ value; the selection standard is the minimum standard. By adjusting different parameters (λ), the binomial deviation of the model is minimized, and the feature datasets with the best performance are selected. (B) Plots the coefficients of the log (λ) function. The λ value is the smallest at the dotted line. Select the coefficient that is not 0 here as the coefficient of the last reserved feature. (C) The y-axis shows the 14 feature names with non-zero coefficients retained at the minimum value of λ, and the x-axis shows their total coefficients in the lasso Cox analysis. The larger the coefficients are, the greater the predictive significance.

Figure 4

The Rad-score of each patient in the training set (A) and validation set (B). The Rad-score is classified according to the threshold value. The Wilcoxon test was used to assess the difference between the two sets.

Figure 5

Radiomics signature ROC curves used to assess predictive performance. (A) The AUC of the training set is 0.86. (B) The AUC of the validation set is 0.82.

Figure 6

Radiomics nomogram for predicting SCLC and NSCLC.

Figure 7

Calibration curves of the radiomics nomogram in the training set (A) and validation set (B). The calibration curves show the calibration of the nomogram in terms of agreement between the predicted probability of SCLC and pathological findings. The 45° blue line indicates perfect prediction, and the dotted lines indicate the predictive performance of the nomogram. The closer the dotted line fit to the ideal line, the better the predictive accuracy of the nomogram.

Figure 8

The AUC was used to estimate the predictive power of different models (A: training set; B: validation set). The radiomics signature and clinical model can be used for the classification of SCLC and NSCLC. In the validation set, the predictive ability of the nomogram (red, AUC = 0.94) was better than that of the clinical model (green, AUC = 0.86). The addition of clinical features improves the prediction efficiency of the radiomics signature.

Figure 9

DCA for the radiomics nomogram. The y-axis shows the net benefit. The red line represents the radiomics nomogram. The blue line indicates the hypothesis that all patients had small cell lung cancer. The black line represents the hypothesis that no patients had small cell lung cancer. The x-axis shows the threshold probability, which is where the expected benefit of treatment is equal to the expected benefit of not undergoing treatment. The decision curves indicate that when the threshold probability is between 0.1 and 1, using the radiomics nomogram to predict small cell lung cancer adds more benefit than treating either all or no patients.