Related Experiment Video
Updated: Apr 12, 2026
10:28
Gene Regulation and Targeted Therapy in Gastric Cancer Peritoneal Metastasis: Radiological Findings from Dual Energy CT and PET/CT
Published on: January 22, 2018
11.1K
Preoperative Risk Stratification for Gastric Cancer: The Establishment of Dual-Energy CT-Based Radiomics Using
Jing Li1, Hongkun Yin2, Huiling Zhang2
1Department of Radiology, The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Zhengzhou 450008, China.
Academic Radiology
|May 11, 2024
Summary
Dual-energy CT (DECT) radiomics models effectively identify high-risk gastric cancer phenotypes, including serosal invasion and lymph node metastasis. These models show promise in preoperative stratification and are linked to disease-free survival.
Area of Science:
- Radiology
- Oncology
- Medical Imaging
Background:
- Gastric cancer diagnosis and staging are critical for treatment planning.
- Accurate identification of high-risk histopathologic phenotypes is essential for predicting patient outcomes.
- Conventional imaging methods have limitations in precisely characterizing these phenotypes.
Purpose of the Study:
- To evaluate dual-energy CT (DECT)-based radiomics models for identifying high-risk histopathologic phenotypes in gastric cancer.
- To assess the performance of DECT radiomics in detecting serosal invasion (pT4a), lymph node metastasis (LNM), lymphovascular invasion (LVI), and perineural invasion (PNI).
- To compare DECT radiomics models with conventional contrast-enhanced CT (CECT) models and clinical models.
Main Methods:
- A prospective bi-center study included 503 patients with gastric adenocarcinoma undergoing DECT before gastrectomy.
- Radiomics features were extracted from various DECT image types and phases.
- Predictive models were developed using logistic regression and validated externally, with performance assessed by AUCs.
Main Results:
- DECT radiomics models achieved high AUCs for identifying pT4a (0.891), LNM (0.817), LVI (0.834), and PNI (0.889) in the validation dataset.
- DECT models demonstrated improved performance over CECT models for pT4a, LNM, and LVI in the training dataset.
- DECT models were associated with improved patient disease-free survival.
Conclusions:
- DECT-based radiomics can effectively stratify gastric cancer patients preoperatively based on high-risk phenotypes.
- These models offer comparable or superior performance to conventional CT methods.
- DECT radiomics show potential for improving prognostic assessment and guiding treatment decisions in gastric cancer.
