Keywords: Prostate, Prostate, Relaxometry, AI/ML Image Reconstruction, Luminal Water Fraction, Deep Learning, WGAN, T2 Decay, Prostate Cancer Imaging
Motivation: Traditional Luminal Water Fraction (LWF) estimation models, such as NNLS, are prone to noise and may underestimate LWF in low-value regions.
Goal(s): This study aims to improve LWF estimation by integrating a Wasserstein Generative Adversarial Network (WGAN) and an artificial neural networks (ANN).
Approach: The ANN-WGAN model generates training data that closely matches real T2 decay signals, which is used to train the ANN for more reliable LWF predictions. Performance is compared with NNLS-based models.
Results: ANN-WGAN outperforms NNLS-based model in simulations, volunteer scans, and patient data, offering enhanced accuracy and robustness in LWF estimation.
Impact: This work introduces a more robust method for Luminal Water Fraction (LWF) estimation in prostate MRI, enhancing lesion differentiation and reducing reliance on predefined training data. It improves LWF consistency and reliability, potentially aiding in prostate cancer imaging and diagnosis.
How to access this content:
For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.
After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.
After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.
Keywords