Project

AI4PKD

AI-based methods to improve stratification of patients affected by Polycystic Kidney Disease using multi-parametric MRI
Application Areas
Research Areas
Research Lines
Staff
WEBSITE

ITB Principal Investigator

Name

AI-based methods to improve stratification of patients affected by Polycystic Kidney Disease using multi-parametric MRI

Acronym

AI4PKD

Location

Segrate

Start Date

2023

End Date

2026

Funder

Ministero dell'Università e della Ricerca - Progetti PRIN2022

Partners

Università di Bergamo, Istituto di Ricerche Farmacologiche Mario Negri IRCCS

Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a hereditary disorder characterized by progressive cyst formation in the kidneys and liver, leading to chronic kidney injury. Total kidney, cyst, and non-cystic parenchyma volumes are key MRI-derived biomarkers for monitoring disease progression and fibrosis. Multi-parametric MRI (mp-MRI) enables non-invasive quantification and characterization of these components. AI4PKD aims to develop robust AI-based methods for automatic segmentation of kidneys, liver, and cysts from anatomical and diffusion MRI. Using data from 35 ADPKD patients and 16 healthy subjects, Bayesian neural networks and explainable AI tools will be trained to enhance reliability. Radiomic analysis of identified structures will support novel biomarkers and stratification models for ADPKD staging.
  • Xiong Q, He X, Scalco E, et al. Automatically Measuring Kidney, Liver and Cyst Volumes in ADPKD. Journal of American Society of Nephrology. 2025. https://doi.org/10.1681/ASN.0000000904
  • Damiano R, Merli A, Lanzarone E, Scalco E. Segmentation Variability in Bayesian U-Net versus Manual Annotations: Impact on Radiomic Reproducibility in Lung Tumor CT Images. Proceedings of the IEEE-EMBS 47th Annual International Conference. Copenhagen Denmark, July 14-17, 2025. https://doi.org/10.1109/EMBC58623.2025.11253060
  • Damiano R, Scalco E, Della Vedova ML, et al.. Integrating Uncertainty Into U-Net Robustness Evaluation Under Natural MRI Alterations: Application to Kidney Segmentation. In: Bellazzi, R., Juarez Herrero, J.M., Sacchi, L., Zupan, B. (eds) Artificial Intelligence in Medicine. AIME 2025. Lecture Notes in Computer Science, vol 15735. Springer, Cham. https://doi.org/10.1007/978-3-031-95841-0_23
  • Damiano R, Lanzarone E, Lussana F, et al. The Impact of Uncertainty Estimation on Radiomic Segmentation Reproducibility and Scan-Rescan Repeatability in Kidney MRI. Medical Physics. 2025. 52 (7), e17995. https://doi.org/10.1002/mp.17995
  • Lussana F, Lanzarone E, Villa G, et al. Reliability of radiomic analysis on multiparametric MRI for patients affected by Autosomal Dominant Polycystic Kidney Disease. Scientific Reports. 2025. 15, 16526. https://doi.org/10.1038/s41598-025-99982-x
  • Scalco E, Pozzi S, Rizzo G, Lanzarone E. Uncertainty quantification in multi-class segmentation: Comparison between Bayesian and non-Bayesian approaches in a clinical perspective. Medical Physics. 2024;1-13. https://doi.org/10.1002/mp.17189