Related Experiment Video
Updated: Sep 27, 2026

Gene Regulation and Targeted Therapy in Gastric Cancer Peritoneal Metastasis: Radiological Findings from Dual Energy CT and PET/CT
Published on: January 22, 2018
Delta 68Ga-Pentixafor PET/CT Radiomics Models for the Detection of Surgically Curable Primary Aldosteronism
Rui Zuo1, Lu Xu1, Tingting Xu2,3
1Department of Nuclear Medicine, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China.
Objective:
Unilateral primary aldosteronism (PA) can be cured through adrenalectomy. We developed a 68Ga-Pentixafor positron emission tomography/computed tomography (PET/CT) radiomics model to detect surgically curable PA.
Methods:
A total of 359 patients with PA who underwent dual-time-phase (10 min and 40 min) 68Ga-Pentixafor PET/CT were enrolled from two centers (center 1 = 268; center 2 = 91). Radiomics models were developed based on 10-min PET, 40-min PET, and delta PET features. The performance of the three PET models and the combined clinical-radiomics model was evaluated using receiver operating characteristic curves. Model accuracy and clinical utility were assessed with calibration curves and decision curve analysis.
Results:
Among the eight radiomics models, the XGBoost model achieved the best classification performance and was selected for the final model. Among the three PET models, the delta PET model yielded the highest area under the curve in the training set (0.875; 95% confidence interval [CI] 0.814-0.935), validation set (0.861; 95% CI 0.754-0.968), and external test set (0.860; 95% CI 0.773-0.946). The combined clinical-radiomics model showed superior area under the curve values: 0.926 (95% CI 0.891-0.962) in training, 0.909 (95% CI 0.836-0.983) in validation, and 0.919 (95% CI 0.855-0.982) in external testing. The combined model outperformed the PET/CT model across all datasets and demonstrated good calibration and clinical utility.
Conclusion:
The model incorporating delta 68Ga-Pentixafor PET/CT radiomics features and clinical features optimizes the selection of patients with surgically curable PA, providing reliable decision support for precision surgical treatment.
