# 3D-StyleGAN2-ADA: Volumetric Synthesis of Realistic Prostate T2W MRI

**Authors:** Claudia Giardina, Verónica Vilaplana

PMC · DOI: 10.3390/jimaging12030130 · Journal of Imaging · 2026-03-14

## TL;DR

This paper introduces a 3D version of StyleGAN2-ADA for generating realistic prostate T2W MRI images at high resolution, showing improved quality and usefulness for medical tasks.

## Contribution

The paper presents a stable 3D StyleGAN2-ADA model for high-resolution prostate MRI synthesis, outperforming prior methods in quality and anatomical accuracy.

## Key findings

- 3D-StyleGAN2-ADA achieved a FID of 27.3, significantly lower than the baseline 3D-StyleGAN's 114.2.
- Synthetic data augmentation matched real-data performance in prostate segmentation tasks.
- Radiomic analyses showed strong alignment between real and synthetic MRI volumes.

## Abstract

This work investigates the extension of StyleGAN2-ADA to three-dimensional prostate T2-weighted (T2W) MRI generation. The architecture is adapted to operate on 3D anisotropic volumes, enabling stable training at a clinically relevant resolution of 256×256×24, where a baseline 3D-StyleGAN fails to converge. Quantitative evaluation using Fréchet Inception Distance (FID), Kernel Inception Distance (KID), and generative Precision–Recall metrics demonstrates substantial improvements over a 3D-StyleGAN baseline. Specifically, FID decreased from 114.2 to 27.3, while generative Precision increased from 0.22 to 0.82, indicating markedly improved fidelity and alignment with the real data distribution. Beyond generative metrics, the synthetic volumes were evaluated through radiomic feature analysis and downstream prostate segmentation. Synthetic data augmentation resulted in segmentation performance comparable to real-data training, supporting that volumetric generation preserves anatomically relevant structures, while multivariate radiomic analyses showed strong global feature alignment between real and synthetic volumes. These findings indicate that a 3D extension of StyleGAN2-ADA enables stable high-resolution volumetric prostate MRI synthesis while preserving anatomically coherent structure and global radiomic characteristics.

## Linked entities

- **Diseases:** prostate cancer (MONDO:0005159)

## Full-text entities

- **Genes:** ADA (adenosine deaminase) [NCBI Gene 100] {aka ADA1}, GAN (gigaxonin) [NCBI Gene 8139] {aka GAN1, GIG, KLHL16}
- **Diseases:** significant (MESH:D065309), injury to (MESH:D014947), Prostate Cancer (MESH:D011471), Alzheimer's (MESH:D000544), cancer (MESH:D009369)
- **Chemicals:** 3D-StyleGAN (-)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC13028575/full.md

## Figures

6 figures with captions in the complete paper: https://tomesphere.com/paper/PMC13028575/full.md

## References

37 references — full list in the complete paper: https://tomesphere.com/paper/PMC13028575/full.md

---
Source: https://tomesphere.com/paper/PMC13028575