Machine learning-based radiomics strategy for prediction of cell proliferation in non-small cell lung cancer. Issue 118 (September 2019)

• Record Type: Journal Article
• Authors: Gu, Qianbiao; Feng, Zhichao; Liang, Qi; Li, Meijiao; Deng, Jiao; Ma, Mengtian; Wang, Wei; Liu, Jianbin; Liu, Peng; Rong, Pengfei
• Is Part Of: European journal of radiology. Issue 118(2019)
• Page Start: 32
• Page End: 37
• Publication Date: 2019-09
• Subjects: Non-small cell lung cancer (NSCLC) -- Ki-67 -- CT -- Radiomics -- Machine learning; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2019.06.025
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: Lung cancer is the leading cause of cancer-related deaths worldwide. Ki-67 nucleoprotein is a key marker that is associated with cell proliferation and tumour heterogeneity. Advanced radiomic features can capture more information than subjective imaging features. Machine learning-based CT radiomics classifiers can noninvasively predict the expression level of Ki-67 in NSCLC. Random forest classifier can produce good classification results. Abstract: Purpose: To explore the feasibility and performance of machine learning-based radiomics classifier to predict the cell proliferation(Ki-67)in non-small cell lung cancer (NSCLC). Methods: 245 histopathological confirmed NSCLC patients who underwent CT scans were retrospectively included. The Ki-67 proliferation index (Ki-67 PI) were measured within 2 weeks after CT scans. A lesion volume of interest (VOI) was manually delineated and radiomics features were extracted by MaZda software from CT images. A random forest feature selection algorithm (RFFS) was used to reduce features. Six kinds of machine learning methods were used to establish radiomics classifiers, subjective imaging feature classifiers and combined classifiers, respectively. The performance of these classifiers was evaluated by the receiver operating characteristic curve (ROC) and compared with Delong test. Results: 103 radiomics features were extracted and 20 optimal features were selected using RFFS. Among the radiomics classifiers established by six machine learning methods, random forest-based radiomics classifier achieved the best performance (AUC = 0.776) in predicting the Ki-67 expression level with sensitivity and specificity of 0.726 and 0.661, which was better than that of subjective imaging classifiers (AUC = 0.625, P < 0.05). However, the combined classifiers did not improve the predictive performance (AUC = 0.780, P > 0.05), with sensitivity and specificity of 0.752 and 0.633. Conclusions: The machine learning-based CT radiomics classifier in NSCLC can facilitate the prediction of the expression level of Ki-67 and provide a novel non-invasive strategy for assessing the cell proliferation.