Whole-esophageal radiomics and deep learning for predicting the pT and pN categories of resectable esophageal cancer
- 1The Affiliated Huai’an No.1 People’s Hospital of Nanjing Medical University
- 2Taizhou First People’s Hospital
- 3Shanghai United Imaging Healthcare Company
Abstract
To investigate the application of whole-esophageal radiomics and deep learning features derived from enhanced arterial phase CT images along with clinical data to differentiate the pT and pN categories in patients with resectable esophageal cancer. Clinical and CT data of 625 patients from Center 1 and 56 patients from Center 2 with esophageal cancer were analyzed. The dataset of Center 1 was randomly divided into a training cohort and an internal testing cohort at an 8:2 ratio. The whole-esophageal ROIs were delineated semi-automatically. The least absolute shrinkage and selection operator algorithm was employed for radiomics and deep learning feature reduction and selection, leading to the development of Radscore and Deepscore. Clinical, radiomics, deep learning, and combined models were constructed. The diagnostic performance of the models was evaluated using the receiver operating characteristic (ROC) curve. In the training cohort for predicting pT categories, the AUC of the combined model was 0.901, which was significantly higher than that of the clinical (0.869), radiomics (0.833), and deep learning (0.830) models. For predicting pN categories, the AUC of the combined model was 0.800, significantly higher than that of the clinical (0.748), radiomics (0.748), and deep learning (0.764) models. Calibration curves also demonstrated good agreement between predicted and actual status, and the decision curve analysis further confirmed its clinical applicability. The combined model offers the most reliable method for predicting thepT and pN categories in patients with resectable esophageal cancer, thereby beneficial to treatment planning.
Related
- The Evolution of Neoadjuvant Treatment for Resectable Esophageal Squamous Cell Carcinoma: Past Achievements, Current Standards, and Future Perspectives
- Efficacy comparison of neoadjuvant chemotherapy, chemoradiotherapy, and chemoimmunotherapy in resectable esophageal squamous cell carcinoma: short- and long-term outcomes from a propensity score-matched analysis
- Neutrophil to lymphocyte ratio (NLR) predicts pathological response to neoadjuvant immunochemotherapy in resectable esophageal cancer: a retrospective study.
- Combined apparent diffusion coefficient histogram and MRI-based fractal analysis for predicting lymph node metastasis in esophageal squamous cell carcinoma
- Association of trilaciclib use with hematologic outcomes during perioperative immunochemotherapy in resectable esophageal squamous cell carcinoma: a prospective observational cohort study
BibTeX
@article{han2026wholeesophageal,
title = {{Whole-esophageal radiomics and deep learning for predicting the pT and pN categories of resectable esophageal cancer}},
author = {Dong Han and Fen Wang and Yang Lu and Xiang Qin and Xin Dai and Yan Zeng and Lili Guo},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62500-8},
url = {https://doi.org/10.1038/s41598-026-62500-8},
note = {CC BY},
}
RIS
TY - JOUR
TI - Whole-esophageal radiomics and deep learning for predicting the pT and pN categories of resectable esophageal cancer
AU - Dong Han
AU - Fen Wang
AU - Yang Lu
AU - Xiang Qin
AU - Xin Dai
AU - Yan Zeng
AU - Lili Guo
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62500-8
UR - https://doi.org/10.1038/s41598-026-62500-8
AB - To investigate the application of whole-esophageal radiomics and deep learning features derived from enhanced arterial phase CT images along with clinical data to differentiate the pT and pN categories in patients with resectable esophageal cancer. Clinical and CT data of 625 patients from Center 1 and 56 patients from Center 2 with esophageal cancer were analyzed. The dataset of Center 1 was randomly divided into a training cohort and an internal testing cohort at an 8:2 ratio. The whole-esophageal ROIs were delineated semi-automatically. The least absolute shrinkage and selection operator algorithm was employed for radiomics and deep learning feature reduction and selection, leading to the development of Radscore and Deepscore. Clinical, radiomics, deep learning, and combined models were constructed. The diagnostic performance of the models was evaluated using the receiver operating characteristic (ROC) curve. In the training cohort for predicting pT categories, the AUC of the combined model was 0.901, which was significantly higher than that of the clinical (0.869), radiomics (0.833), and deep learning (0.830) models. For predicting pN categories, the AUC of the combined model was 0.800, significantly higher than that of the clinical (0.748), radiomics (0.748), and deep learning (0.764) models. Calibration curves also demonstrated good agreement between predicted and actual status, and the decision curve analysis further confirmed its clinical applicability. The combined model offers the most reliable method for predicting thepT and pN categories in patients with resectable esophageal cancer, thereby beneficial to treatment planning.
SN - 2045-2322
ER -