Fig. 1: Overview of the study.
From: Large-scale generative tumor synthesis in computed tomography images for improving tumor recognition
a We explore tumor synthesis and segmentation on five types of tumors/lesions, i.e., liver tumors, pancreas tumors, kidney tumors, lung tumors, and COVID-19. b The rapid advancements in medical imaging have enabled the collection of large-scale Computed Tomography (CT) data. However, annotated tumor datasets are scarce due to the extensive annotation burden. c We curated 161,310 CT volumes from 33 public sources to enable large-scale tumor synthesis and recognition, with merely 2.3% of them comprising annotated tumors. d FreeTumor consists of two stages: synthesis training and segmentation training. In Stage 1, FreeTumor effectively unleashes the power of large-scale unlabeled data for tumor synthesis training. In Stage 2, FreeTumor synthesizes high-quality tumors on healthy organs, facilitating the integration of large-scale unlabeled data in tumor segmentation training. We present two lung instances to demonstrate that we synthesize both lung tumors and COVID-19 lesions on lungs. e Clinical evaluation of synthetic tumors. We invited 13 board-certified radiologists to a Visual Turing Test to discern between synthetic and real tumors. Rigorous clinician evaluation validates the high quality of our synthetic tumors. f Extensive segmentation results on 12 public datasets showcase the superiority of FreeTumor. Specifically, FreeTumor adopts SwinUNETR51 as the segmentation model and employs tumor synthesis for augmenting segmentation datasets. With large-scale synthetic tumors for training, FreeTumor surpasses the baseline SwinUNETR51 by significant margins, achieving 10.6%, 5.5%, 3.8%, 6.1%, and 7.9% Dice score improvements for five types of tumors/lesions, respectively. g Early tumor detection results on 12 public datasets (number of samples n = 1533). Box plots show the mean (center), 25th and 75th percentiles (bounds of box), and minima to maxima (whiskers). With tumor synthesis, FreeTumor yields + 16.4% sensitivity improvements on average. Source data are provided as a Source Data file. The elements are created in BioRender. Wu, L. (2025) https://BioRender.com/qo600iw.
