An artificial intelligence generated automated algorithm to measure total kidney volume in ADPKD

Introduction Accurate tools to inform individual prognosis in patients with autosomal dominant polycystic kidney disease (ADPKD) are lacking. Here, we report an artificial intelligence (AI) generated method for routinely measuring total kidney volume (TKV). Methods An ensemble U-net algorithm was created using the nnUNet approach. The training and internal cross-validation cohort consisted of all 1.5T MRI data acquired using 5 different MRI scanners (454 kidneys, 227 scans) in the CYSTic consortium which was first manually segmented by a single human operator. As an independent validation cohort, we utilised 48 sequential clinical MRI scans with reference results of manual segmentation acquired by 6 individual analysts at a single centre. The tool was then implemented for clinical use and its performance analysed. Results The training / internal validation cohort was younger (mean age 44.0 vs 51.5 years) and the female-male ratio higher (1.2 v 0.94) compared to the clinical validation cohort. The majority of CYSTic patients had PKD1 mutations (79%) and typical disease (Mayo Imaging Class 1, 86%). The median DICE score on the clinical validation dataset between the algorithm and human analysts was 0.96 for left and right kidneys with a median TKV error of -1.8%. The time taken to manually segment kidneys in the CYSTic dataset was 56 (±28) min whereas manual corrections of the algorithm output took 8.5 (±9.2) min per scan. Conclusions Our AI-based algorithm demonstrates performance comparable to manual segmentation. Its rapidity and precision in real-world clinical cases demonstrate its suitability for clinical application.