← Back to all news
arXiv Computer Vision By Xavier Rafael-Palou (Quantitative Imaging Biomarkers in Medicine), Alejandro Vergara-Richart (Quantitative Imaging Biomarkers in Medicine, Universitat Polit\`ecnica de Val\`encia, Val\`encia, Spain), Ignacio Iborra Roncales (Quantitative Imaging Biomarkers in Medicine), Carmen Prieto-de-la-Lastra (Quantitative Imaging Biomarkers in Medicine), Andreu Antolin-Redondo (Department of Radiology, Institut de Diagn\`ostic per la Imatge, Department of Surgery, Universitat Aut\`onoma de Barcelona, Bellaterra, Spain), Juan Morote (Department of Surgery, Universitat Aut\`onoma de Barcelona, Bellaterra, Spain), Ana Jimenez-Pastor (Quantitative Imaging Biomarkers in Medicine)

Conditional Diffusion-Based Generation of T2-Weighted Prostate MRI for Clinically Significant Prostate Cancer Classification

Read the original on arXiv Computer Vision →

The study presents a conditional diffusion-based generative model that augments T2-weighted prostate MRI data with clinically significant prostate cancer (csPCa) and transition zone (TZ) lesions. By extending MAISI‑v2 with csPCa class, zonal, and lesion‑mask conditioning and fine‑tuning on 1,303 patients, the model achieves a substantial reduction in Fréchet inception distance (FID) from 13.28 to 2.04. Synthetic augmentation improves csPCa classification AUC across five simulated training scenarios, though the benefit does not significantly transfer to a cross‑institutional dataset.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Computer Vision
Oct 23

Diffusion-Based Synthesis of Complete 3D Biparametric Prostate MRI

The paper introduces a conditional 3D diffusion framework that synthesizes complete biparametric prostate MRI (bpMRI) from anatomical masks. It supports both sequential synthesis—generating T2-weighted and diffusion modalities with separate models—and joint synthesis—producing all bpMRI modalities simultaneously in a single model. Experiments on PI-CAI, FastMRI Prostate, and PROSTATEx datasets show that joint synthesis achieves competitive distributional fidelity, while directly synthesizing ADC and high‑b‑value DWI outperforms deriving them from synthetic diffusion-weighted images.

By Claudia Giardina, Montse Pard\`as, Ver\'onica Vilaplana
diffusion
More like this →
arXiv Computer Vision
Oct 27

End-to-end prostate lesion segmentation and detection from undersampled diffusion-weighted MRI

arXiv:2610.33554v1 Announce Type: new Abstract: Automated prostate lesion detection from diffusion-weighted MR imaging (DWI) conventionally relies on reconstructed images and does not fully use all a...

By Lukas T. Rotkopf, Marco Schlimbach, Kevin Kr\"{o}ninger, Stefan O. Schoenberg, Heinz-Peter Schlemmer, Jens Kleesiek, Moritz Rempe
diffusioncomputer-vision
More like this →