A pairwise radiomics algorithm - lesion pair relation estimation (PRE) model for distinguishing multiple primary lung cancer (MPLC) from intrapulmonary metastasis (IPM)

作者：Ting-Fei Chen, Lei Yang, Hai-Bin Chen, Hao‐Shuai Yang, Zhen-Tian Wu, Honghe Luo, Qiong Li, Ying Zhu · 发表于：Research Square · 年份：2022 · DOI：10.21203/rs.3.rs-2216255/v1 · 研究领域：Radiomics and Machine Learning in Medical Imaging、Lung Cancer Diagnosis and Treatment、Medical Imaging Techniques and Applications

Abstract Background Distinguishing multiple primary lung cancer (MPLC) from intrapulmonary metastasis (IPM) is critical for their disparate treatment strategy and prognosis. This study aimed to establish a non-invasive model to make the differentiation pre-operatively. Methods We retrospectively studied 168 patients with multiple lung cancers (307 pairs of lesions) including 118 cases for modeling and internal validation, and 50 cases for independent external validation. Radiomic features on computed tomography (CT) were extracted to calculate the absolute deviation of paired lesions. Features were then selected by correlation coefficients and random forest classifier five-fold cross-validation, based on which the lesion pair relation estimation (PRE) model was developed. A major voting strategy was used to decide diagnosis for cases with multiple pairs of lesions. Cases from another institute were included as the external validation set for the PRE model to compete with two experienced clinicians. Results Seven radiomic features were selected for the PRE model construction. With major voting strategy, the mean area under receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the training vs. internal validation vs. external validation cohort to distinguish MPLC were 0.983 vs. 0.844 vs. 0.793, 0.942 vs. 0.846 vs. 0.760, 0.905 vs. 0.728 vs. 0.727, and 0.962 vs. 0.910 vs. 0.769, respectively. AUCs of the two clinicians were 0.619 and 0.580. Con...