A new biomarker combining multimodal MRI radiomics and clinical indicators for differentiating inverted papilloma from nasal polyp invaded the olfactory nerve possibly

Abstract

Background and purpose: Inverted papilloma (IP) and nasal polyp (NP), as two benign lesions, are difficult to distinguish on MRI imaging and clinically, especially in predicting whether the olfactory nerve is damaged, which is an important aspect of treatment and prognosis. We plan to establish a new biomarker to distinguish IP and NP that may invade the olfactory nerve, and to analyze its diagnostic efficacy.

Materials and methods: A total of 74 cases of IP and 55 cases of NP were collected. A total of 80% of 129 patients were used as the training set (59 IP and 44 NP); the remaining were used as the testing set. As a multimodal study (two MRI sequences and clinical indicators), preoperative MR images including T2-weighted magnetic resonance imaging (T2-WI) and contrast-enhanced T1-weighted magnetic resonance imaging (CE-T1WI) were collected. Radiomic features were extracted from MR images. Then, the least absolute shrinkage and selection operator (LASSO) regression method was used to decrease the high degree of redundancy and irrelevance. Subsequently, the radiomics model is constructed by the rad scoring formula. The area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the model have been calculated. Finally, the decision curve analysis (DCA) is used to evaluate the clinical practicability of the model.

Results: There were significant differences in age, nasal bleeding, and hyposmia between the two lesions (p < 0.05). In total, 1,906 radiomic features were extracted from T2-WI and CE-T1WI images. After feature selection, using 12 key features to bulid model. AUC, sensitivity, specificity, and accuracy on the testing cohort of the optimal model were, respectively, 0.9121, 0.828, 0.9091, and 0.899. AUC on the testing cohort of the optimal model was 0.9121; in addition, sensitivity, specificity, and accuracy were, respectively, 0.828, 0.9091, and 0.899.

Conclusion: A new biomarker combining multimodal MRI radiomics and clinical indicators can effectively distinguish between IP and NP that may invade the olfactory nerve, which can provide a valuable decision basis for individualized treatment.

Keywords: MRI; diagnostic efficacy; inverted papilloma; nasal polyp; olfactory nerve; radiomic.

Figure 1

Filter flowchart. Through the inclusion and exclusion criteria of each step, a total of 129 patients were enrolled in the study and randomly assigned to training or testing sets.

Figure 2

(A) Flowchart of the study. An overview of the study workflow including image preprocessing, feature extraction, feature selection, and model construction. (B) Some cases of lesions drawn by radiologists. The tumor contour on the CE-T1WI image was determined, and the CE-T1WI image was referred to when sketching the T2-WI image.

Figure 3

Classification of 1906 features, including first order, glcm, gldm, glrlm, glszm, and shape.

Figure 4

LASSO model dimensionality reduction visualization process. (A) Feature dimensionality reduction using LASSO, the horizontal coordinate indicates the value of the penalty coefficient, and the vertical coordinate indicates the change of the binomial deviation size as the value of the coefficient changes. (B) Feature dimensionality reduction using LASSO, the lower horizontal coordinate indicates the log(λ) value, the upper horizontal coordinate indicates the number of features corresponding to log(λ), and the vertical coordinate indicates the number of each feature’s weight coefficient. (C) Coefficient values of different radiomic signature parameters when constructing rad-score after feature dimensionality reduction using LASSO.

Figure 5

(A) ROC curve of 11 models in the test set. Each color represents a model. (B) Confusion matrix for ExtraTrees model in the test set.

Figure 6

DCA for ExtraTrees model in the test set. The predictive model curves are significantly farther away from the two extreme lines, indicating a good overall net benefit in the population.