SOURCE-LINKED INTELLIGENCE

Diffusion-Based Tumor Inpainting for Renal Segmentation under Clinical Data Scarcity

arXiv · AI, language, vision and robotics · article · Sep 18, 2026 · UTC

Deep learning segmentation of renal tumors requires large annotated datasets, yet clinical deployments typically offer only a handful of tumor-positive cases from the target site. We propose a diffusion-based inpainting framework that synthesizes anatomically plausible renal tumors within healthy CT scans, requiring no additional annotation, and provide the first systematic comparison of 2D, 2.5D, and full 3D (MAISI) synthesis strategies for this task. Training the diffusion model on public data (KiTS23, KIRC) and evaluating nnU-Net segmentation on a internal cohort across three low-data regim

Read original source ↗ Open in workspace

recordType

paper

region

Global