Abstract: FR-PO004
ADPKD Predictor: A Cloud-Based Prognostic Tool for Autosomal Dominant Polycystic Kidney Disease
Session Information
- AI, Digital Health, Data Science - II
November 03, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Carbone, Vincenzo, InSilicoTrials Technologies, Trieste, Italy
- Gazzin, Matteo, InSilicoTrials Technologies, Trieste, Italy
- Bursi, Roberta, InSilicoTrials Technologies, Trieste, Italy
- Magistroni, Riccardo, Università di Modena e Reggio Emilia, Modena, Italy
- Corsi, Cristiana, University of Bologna, Bologna, Italy
Background
Total Kidney Volume is accepted by FDA and EMA as a prognostic biomarker for ADPKD and is currently used to select patients eligible for drug treatment. We developed ADPKD Predictor, a user-friendly cloud-based tool for fast and accurate estimation of disease classification and progression, based on advanced image processing techniques.
Methods
The tool was designed on Microsoft Azure Cloud to facilitate the use of a MATLAB algorithm to automatically detect kidneys and cysts contours from MRI data (Figure 1). TKV is automatically calculated and ADPKD Imaging Classification, eGFR, GFR Category, eligibility for drug treatment and estimated effect are obtained (Figure 2).
Results
The proposed solution is extremely fast and precise compared to manual segmentation (absolute mean error 2.4% ± 2.7%), and more accurate than ellipsoid-based method, resulting in a manifold reduction of misclassification error (2.5%). No numerical expertise, software or hardware is required since computations run remotely in the cloud.
Conclusion
ADPKD Predictor provides a fast and reproducible assessment of risk classification and disease progression. It is extremely useful to researchers and clinicians for effective stratification of patients, hence supporting correct therapy administration.
ADPKD Predictor - Input page
ADPKD Predictor - Results page