Opportunistic Screening with Radiomics Model Based on CT Imaging of Adrenal Adenoma to Assess Bone Mass Change

There was a strong correlation between adrenal adenoma and osteoporosis, the primary objective of this research was to establish and authenticate a radiomics nomogram using CT scan of adrenal adenoma to screen abnormal bone mineral density (BMD) opportunistically.

A total of 161 patients with adrenal adenomas who underwent thoracoabdominal CT and quantitative CT (QCT) were enrolled retrospectively. The radiomics features were chosen from the cross-sectional CT images of adrenal adenomas and the nomogram models that including patient's clinical and radiomics features were then established. The receiver operating characteristic (ROC) curve was performed to evaluate the performance of the model and the decision curve analysis (DCA) was used to assess the clinical usefulness.

To build a radiomics model, 11 radiomics features based on CT scans of adrenal adenomas were selected and showed good performance in distinguishing abnormal BMD from normal BMD. Moreover, the radiomics nomogram model demonstrated excellent ability to identify abnormal BMD of adrenal adenoma patients with area under the curve (AUC) of 0.87 (95% CI, 0.80-0.93) in training cohort and 0.85 (95% CI, 0.74-0.96) in validation cohort. The accuracy, sensitivity, specificity of the nomogram model were 79.7%, 78.3%, 81.1% in training cohort, and 72.9%, 67.7%, 82.4% in validation cohort respectively.

The radiomics nomogram based on clinical and radiomics features of adrenal adenoma CT images had a satisfying predictive ability and can be an opportunistic effective tool for identifying bone mass change.