CT imaging-based histogram features for prediction of EGFR mutation status of bone metastases in patients with primary lung adenocarcinoma

Abstract

Objective To identify imaging markers that reflect the epidermal growth factor receptor (EGFR) mutation status by comparing computed tomography (CT) imaging-based histogram features between bone metastases with and without EGFR mutation in patients with primary lung adenocarcinoma. Materials and methods This retrospective study included 57 patients, with pathologically confirmed bone metastasis of primary lung adenocarcinoma. EGFR mutation status of bone metastases was confirmed by gene detection. The CT imaging of the metastatic bone lesions which were obtained between June 2014 and December 2017 were collected and analyzed. A total of 42 CT imaging-based histogram features were automatically extracted. Feature selection was conducted using Student’s t-test, Mann-Whitney U test, single-factor logistic regression analysis and Spearman correlation analysis. A receiver operating characteristic (ROC) curve was plotted to compare the effectiveness of features in distinguishing between EGFR(+) and EGFR(−) groups. DeLong’s test was used to analyze the differences between the area under the curve (AUC) values. Results Three histogram features, namely range, skewness, and quantile 0.975 were significantly associated with EGFR mutation status. After combining these three features and combining range and skewness, we obtained the same AUC values, sensitivity and specificity. Meanwhile, the highest AUC value was achieved (AUC 0.783), which also had a higher sensitivity (0.708) and specificity (0.788). The differences between AUC values of the three features and their various combinations were statistically insignificant. Conclusion CT imaging-based histogram features of bone metastases with and without EGFR mutation in patients with primary lung adenocarcinoma were identified, and they may contribute to diagnosis and prediction of EGFR mutation status. Electronic supplementary material The online version of this article (10.1186/s40644-019-0221-9) contains supplementary material, which is available to authorized users.

Radiomics Analysis of Iodine-Based Material Decomposition Images With Dual-Energy Computed Tomography Imaging for Preoperatively Predicting Microsatellite Instability Status in Colorectal Cancer

“…The ROIs were required to include the area of necrosis and bleeding within the tumor and excluded perienteric fat and intestinal contents. To correct for acquisition-related differences of differing voxel resolutions in the two different CT scanners, voxel dimensions (mm) of each iodine-based MD image dataset were isotropically resampled to a common voxel spacing 0.5 × 0.5 × 0.5 mm 3 ( x, y, z ) via linear interpolation algorithm (18, 19). Next, a total of 606 radiomics features for each CRC patient were extracted via Artificial Intelligent Kit (GE Healthcare) in concordance with the reference manual by the “Image Biomarker Standardization Initiative.” These features were divided into four groups: (1) first-order histogram features ( n = 42); (2) second-order texture features: gray level co-occurrence matrix (GLCM) ( n = 240), Haralick features ( n = 10); (3) grey-level zone size matrix (GLZSM) ( n = 11); and (4) Gaussian transform ( n = 303).…”

Application Value of Whole-Body Pet/ct and Contrast-enhanced Brain Mri in the Diagnosis and Treatment of Patients With Synchronous Multiple Pulmonary Ground-glass Nodules

“…All scanning was performed on a Gemini GXL 16-slice PET/CT system (Philips) with 18 Fuorodeoxyglucose (FDG) (radiochemical identity/purity > 95%) provided by Andico. The patient had fasted for more than 6 hours.…”

“…The patient's height and weight and level of fasting blood glucose (<6.1 mmol/L) were measured. In the resting state, 222-492.1 MBq (6-13.3 mCi) 18 F-FDG was injected via the dorsal vein of the hand. PET/CT was performed 50-60 minutes after injection, during which the patient was resting in a dark room.…”

A Machine Learning Model Based on PET/CT Radiomics and Clinical Characteristics Predicts ALK Rearrangement Status in Lung Adenocarcinoma

“…A large number of studies have shown that the radiomic analysis can be used to predict the mutation status of several oncogenes (16,17). Currently, most studies in lung cancer have been done in primary tumors using computed tomography (CT) images (18)(19)(20)(21)(22). For example, Gevaert et al used CT images-based signature of primary lung tumors to predict EGFR mutation status (23).…”