KidMesh: Computational Mesh Reconstruction for Pediatric Congenital Hydronephrosis Using Deep Neural Networks

TL;DR

This work introduces KidMesh, an end-to-end explicit deep-learning framework that reconstructs CFD-ready meshes of pediatric hydronephrosis directly from MRU images. It integrates a three-module pipeline—Feature Extraction (FEM), Feature Sampling (FSM), and Mesh Deformation (MDM)—guided by a weakly supervised training regime that leverages pseudo-gold standard meshes derived from segmentations. Key innovations include a dynamic vertex upsampling strategy, a learnable neighborhood sampling with self-attention, and topology-preserving regularizers that ensure watertight, high-quality meshes suitable for urodynamic CFD simulations. Extensive experiments show that KidMesh delivers fast, accurate, and topologically robust meshes with comparable functional CFD results to manual references, enabling rapid, patient-specific analysis of urinary dynamics. The framework demonstrates strong potential for clinical deployment and provides a foundation for extending CFD-ready mesh reconstruction to the broader urinary tract.

Abstract

Pediatric congenital hydronephrosis (CH) is a common urinary tract disorder, primarily caused by obstruction at the renal pelvis-ureter junction. Magnetic resonance urography (MRU) can visualize hydronephrosis, including renal pelvis and calyces, by utilizing the natural contrast provided by water. Existing voxel-based segmentation approaches can extract CH regions from MRU, facilitating disease diagnosis and prognosis. However, these segmentation methods predominantly focus on morphological features, such as size, shape, and structure. To enable functional assessments, such as urodynamic simulations, external complex post-processing steps are required to convert these results into mesh-level representations. To address this limitation, we propose an end-to-end method based on deep neural networks, namely KidMesh, which could automatically reconstruct CH meshes directly from MRU. Generally, KidMesh extracts feature maps from MRU images and converts them into feature vertices through grid sampling. It then deforms a template mesh according to these feature vertices to generate the specific CH meshes of MRU images. Meanwhile, we develop a novel schema to train KidMesh without relying on accurate mesh-level annotations, which are difficult to obtain due to the sparsely sampled MRU slices. Experimental results show that KidMesh could reconstruct CH meshes in an average of 0.4 seconds, and achieve comparable performance to conventional methods without requiring post-processing. The reconstructed meshes exhibited no self-intersections, with only 3.7% and 0.2% of the vertices having error distances exceeding 3.2mm and 6.4mm, respectively. After rasterization, these meshes achieved a Dice score of 0.86 against manually delineated CH masks. Furthermore, these meshes could be used in renal urine flow simulations, providing valuable urodynamic information for clinical practice.

Figures (10)

• Figure 1: Illustration of voxel-based magnetic resonance urography (MRU) images and mesh reconstruction methods. The left image shows a patch from MRU images, with the yellow contour indicating hydronephrosis boundaries. MRU techniques visualize this region using voxels, often resulting in staircase artifacts (black circles) after reconstruction. The right top panel displays meshes generated using Marching Cubes with post-processing, while the right bottom panel shows our proposed reconstruction meshes based on deep learning.
• Figure 2: Workflow of the three mainstream deep learning mesh reconstruction methods: voxel-based, implicit, and explicit. The first two methods necessitate complex post-processing steps, such as isosurface extraction and topology correction (TC), following image segmentation (e.g., region-growing algorithms) or function mapping (e.g., signed distance function and occupancy function). In contrast, the explicit reconstruction method, which employs CNN and GCN, generates high-precision meshes while preserving the topology, thereby eliminating the need for post-processing.
• Figure 3: Overall architecture of KidMesh. The network takes MRU images and a mesh template as input and generates computational meshes for congenital hydronephrosis in a coarse-to-fine manner. The orange box denotes the Feature Extraction Module (FEM), which follows a U-shaped encoder–decoder topology to extract hierarchical 3D feature maps. The blue box represents the Feature Sampling Module (FSM), which samples vertex-wise features via grid sampling. The purple box depicts the Mesh Deformation Module (MDM), which progressively deforms the mesh template into the target shape under multi-scale feature guidance.
• Figure 4: The illustration of vertex upsampling, consists of uniform unpool (UU) and vertex filter (VF). UU aims to enlarge the number of vertex in generated meshes, while VF filters out no and negative contribution vertices.
• Figure 5: Visualization of the reconstructed meshes of KidMesh with different configurations. Black circles highlight the regions where have significant mesh differences.