A Three-Dimensional Skeletonization Algorithm for Morphological Analysis of Renal Cysts in Patients with Autosomal Dominant Polycystic Kidney Disease

Abstract

Autosomal dominant polycystic kidney disease (ADPKD) is a disease in which multiple cysts are continuously developed and enlarged in the renal parenchyma over the lifetime. It is important and yet challenging to provide a useful index to predict a prognostic pattern for renal dysfunction at the early stage of the disease. This study aimed to propose a 3D skeletonization algorithm so as to extract the morphological features of the 3D shape of the renal cysts. Ten sets of T2-weighted abdominal MR images were processed and the cumulative relative frequency of the extracted skeletonsʼ length was compared with a conventional renal functional index: eGFR. The skeletons could quantify the distribution of the size of renal cysts, and the relative frequency of short skeleton (≤3 voxels) corresponded to the eGFR. The proposed method would be valuable to assess a progression pattern in the growth and occurrence of the renal cysts in patients with ADPKD.