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

2026-02-09

AI Summary

This work proposes KidMesh, an end-to-end deep learning framework that directly reconstructs high-quality, self-intersection-free, and geometrically accurate 3D mesh models of hydronephrotic kidneys from pediatric magnetic resonance urography (MRU) images, circumventing the limitations of conventional voxel-based segmentation methods that require complex post-processing and fail to produce simulation-ready meshes. KidMesh operates without precise mesh annotations by leveraging feature-point-guided template deformation, integrating MRU-specific feature extraction with a tailored mesh sampling strategy for efficient training. Experimental results demonstrate that the method reconstructs meshes in under 0.4 seconds on average, with 96.3% of vertices exhibiting errors below 3.2 mm; rasterized outputs achieve a Dice coefficient of 0.86 against manual annotations, substantially facilitating downstream urodynamic simulations.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

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.

Problem

• congenital hydronephrosis
• mesh reconstruction
• magnetic resonance urography
• urodynamic simulation
• computational mesh

Innovation

• mesh reconstruction
• deep neural networks
• pediatric congenital hydronephrosis
• template deformation
• annotation-free training