AUTHOR=Wang Dongqing , Zhuang Zijian , Wu Shuting , Chen Jixiang , Fan Xin , Liu Mengsi , Zhu Haitao , Wang Ming , Zou Jinmei , Zhou Qun , Zhou Peng , Xue Jing , Meng Xiangpan , Ju Shenghong , Zhang Lirong 

TITLE=A Dual-Energy CT Radiomics of the Regional Largest Short-Axis Lymph Node Can Improve the Prediction of Lymph Node Metastasis in Patients With Rectal Cancer

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.846840

DOI=10.3389/fonc.2022.846840

ISSN=2234-943X

ABSTRACT=Objective: To explore the value of dual-energy computed tomography (DECT) radiomics of the regional largest short-axis lymph nodes for evaluating lymph node staging in patients with rectal cancer.
Materials and Methods: One hundred forty-one patients with rectal cancer (58 in LNM+ group, 83 in LNM- group) who underwent preoperative total abdominal energy spectrum CT were divided into a training group and validation group (7:3 ratio). After post-processing DECT portal vein images, 120kVp-like images and iodine (water) images were obtained. The highest-risk lymph nodes were identified; their long-axis and short-axis diameter and quantitative energy spectrum parameters were measured manually by two experienced radiologists who were blind to the postoperative pathological results. Four DECT parameters were analyzed: arterial phase (AP) standardized iodine uptake value, the arterial phase standardized effective atomic number, the venous phase (VP) standardized iodine uptake value, and the venous  phase standardized effective atomic number. The carcinoembryonic antigen (CEA) index level within one week before surgery was also recorded. Radomics signatures of 120kVp-like images (Rad signature120kVp) and iodine map (Rad signatureImap) were built based on Least Absolute Shrinkage and Selection Operator (LASSO).
Results: Eight hundred thirty-three features (2*833) were extracted from 120Kvp-like and iodine images. In the independent validation group, the radiomics features based on 120kVp-like images showed the best diagnostic performance (AUC=0.922) compared to other models [CT morphological indicators (short-axis diameter (AUC=0.779) and long-axis diameter alone (AUC=0.714), CEA alone (AUC=0.540), and standardized dual-energy CT parameters alone (AUC=0.504-0.718)]. Contrary, DECT iodine map-based radiomic signatures showed no better diagnostic ability (AUC=0.866) in predicting lymph node metastasis. The decision curve showed that the 120kVp-like-based imaging omics label has the highest net income.
Conclusion: Predictive model based on a 120kVp-like image and the largest short-axis diameter lymph nodes has the highest diagnostic value in predicting lymph node metastasis in patients with rectal cancer.