5 papers
PCaPaint: Prostate Cancer Inpainting by Mitigating Shortcut Learning
Levente Lippenszky, Hongxu Yang, Marcell Dömötör +2
The development of AI systems for tumor-specific applications is limited by the scarcity of labeled data. Synthetic tumor inpainting offers a promising approach but faces challenge…
SynthRAR: Ring Artifacts Reduction in CT with Unrolled Network and Synthetic Data Training
Hongxu Yang, Levente Lippenszky, Edina Timko +1
Defective and inconsistent responses in CT detectors can cause ring and streak artifacts in the reconstructed images, making them unusable for clinical purposes. In recent years, s…
Low performing pixel correction in computed tomography with unrolled network and synthetic data training
Hongxu Yang, Levente Lippenszky, Edina Timko +2
Low performance pixels (LPP) in Computed Tomography (CT) detectors would lead to ring and streak artifacts in the reconstructed images, making them clinically unusable. In recent y…
SynBT: High-quality Tumor Synthesis for Breast Tumor Segmentation by 3D Diffusion Model
Hongxu Yang, Edina Timko, Levente Lippenszky +2
Synthetic tumors in medical images offer controllable characteristics that facilitate the training of machine learning models, leading to an improved segmentation performance. Howe…
Modality-Independent Explainable Detection of Inaccurate Organ Segmentations Using Denoising Autoencoders
Levente Lippenszky, István Megyeri, Krisztian Koos +7
In radiation therapy planning, inaccurate segmentations of organs at risk can result in suboptimal treatment delivery, if left undetected by the clinician. To address this challeng…