AUTHOR=Gao Xujie , Cui Jingli , Wang Lingwei , Wang Qiuyan , Ma Tingting , Yang Jilong , Ye Zhaoxiang 

TITLE=The value of machine learning based radiomics model in preoperative detection of perineural invasion in gastric cancer: a two-center study

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1205163

DOI=10.3389/fonc.2023.1205163

ISSN=2234-943X

ABSTRACT=Purpose: To establish and validate a machine learning based radiomics model for detection of perineural invasion (PNI) in gastric cancer (GC).
Methods: This retrospective study included a total of 955 patients with GC selected from two centers; they were separated into training (n=603), internal testing (n=259), and external testing (n=93) sets. Radiomic features were derived from three phases of contrast-enhanced computed tomography (CECT) scan images. A radiomics signature was established with robust features, utilizing the lowest absolute shrinkage and selection operator approach. A radiomics model was constructed by aggregating the radiomic signatures and important clinicopathological characteristics. The predictive ability of the radiomic model was then assessed with receiver operating characteristic (ROC) and calibration curve analyses in all three sets.
Results: The PNI rates for the training, internal testing, and external testing sets were 22.1, 22.8, and 36.6 %, respectively. The radiomics signature, consisting of 9 robust characteristics, revealed good discrimination accuracy for the PNI in all three sets (training set: AUC = 0.85; internal testing set: AUC = 0.81; external testing set: AUC = 0.79). The risk of PNI was significantly associated with higher radiomics scores. A combined model that integrated radiomics and pathological factors demonstrated enhanced accuracy and excellent calibration in all three sets (training set: AUC = 0.89; internal testing set: AUC = 0.85; external testing set: AUC = 0.83).
Conclusion: The suggested radiomics model exhibited satisfactory prediction performance for the PNI in GC.