AI-based segmentation of renal enhanced CT images for quantitative evaluate of chronic kidney disease

To quantitatively evaluate chronic kidney disease (CKD), a deep convolutional neural network-based segmentation model was applied to renal enhanced computed tomography (CT) images. A retrospective analysis was conducted on a cohort of 100 individuals diagnosed with CKD and 90 individuals with healthy kidneys, who underwent contrast-enhanced CT scans of the kidneys or abdomen. Demographic and clinical data were collected from all participants. The study consisted of two distinct stages: firstly, the development and validation of a three-dimensional (3D) nnU-Net model for segmenting the arterial phase of renal enhanced CT scans; secondly, the utilization of the 3D nnU-Net model for quantitative evaluation of CKD. The 3D nnU-Net model achieved a mean Dice Similarity Coefficient (DSC) of 93.53% for renal parenchyma and 81.48% for renal cortex. Statistically significant differences were observed among different stages of renal function for renal parenchyma volume (V_RP), renal cortex volume (V_RC), renal medulla volume (V_RM), the CT values of renal parenchyma (Hu_RP), the CT values of renal cortex (Hu_RC), and the CT values of renal medulla (Hu_RM) (F = 93.476, 144.918, 9.637, 170.533, 216.616, and 94.283; p < 0.001). Pearson correlation analysis revealed significant positive associations between glomerular filtration rate (eGFR) and V_RP, V_RC, V_RM, Hu_RP, Hu_RC, and Hu_RM (r = 0.749, 0.818, 0.321, 0.819, 0.820, and 0.747, respectively, all p < 0.001). Similarly, a negative correlation was observed between serum creatinine (Scr) levels and V_RP, V_RC, V_RM, Hu_RP, Hu_RC, and Hu_RM (r = - 0.759, - 0.777, - 0.420, - 0.762, - 0.771, and - 0.726, respectively, all p < 0.001). For predicting CKD in males, V_RP had an area under the curve (AUC) of 0.726, p < 0.001; V_RC, AUC 0.765, p < 0.001; V_RM, AUC 0.578, p = 0.018; Hu_RP, AUC 0.912, p < 0.001; Hu_RC, AUC 0.952, p < 0.001; and Hu_RM, AUC 0.772, p < 0.001 in males. In females, V_RP had an AUC of 0.813, p < 0.001; V_RC, AUC 0.851, p < 0.001; V_RM, AUC 0.623, p = 0.060; Hu_RP, AUC 0.904, p < 0.001; Hu_RC, AUC 0.934, p < 0.001; and Hu_RM, AUC 0.840, p < 0.001. The optimal cutoff values for predicting CKD in Hu_RP are 99.9 Hu for males and 98.4 Hu for females, while in Hu_RC are 120.1 Hu for males and 111.8 Hu for females. The kidney was effectively segmented by our AI-based 3D nnU-Net model for enhanced renal CT images. In terms of mild kidney injury, the CT values exhibited higher sensitivity compared to kidney volume. The correlation analysis revealed a stronger association between V_RC, Hu_RP, and Hu_RC with renal function, while the association between V_RP and Hu_RM was weaker, and the association between V_RM was the weakest. Particularly, Hu_RP and Hu_RC demonstrated significant potential in predicting renal function. For diagnosing CKD, it is recommended to set the threshold values as follows: Hu_RP < 99.9 Hu and Hu_RC < 120.1 Hu in males, and Hu_RP < 98.4 Hu and Hu_RC < 111.8 Hu in females.