Abstract: PO0490
Automated Tubular Morphometric Analysis in Kidney Biopsies
Session Information
- Bioengineering: Organoids and Organs-on-a-Chip
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Bioengineering
- 300 Bioengineering
Authors
- Ginley, Brandon, University at Buffalo, Buffalo, New York, United States
- Kavthekar, Neil S., University at Buffalo, Buffalo, New York, United States
- Lucarelli, Nicholas, University at Buffalo, Buffalo, New York, United States
- Santo, Briana A., University at Buffalo, Buffalo, New York, United States
- Jen, Kuang-Yu, University of California Davis, Davis, California, United States
- Sarder, Pinaki, University at Buffalo, Buffalo, New York, United States
Background
Tubular atrophy is prevalent in kidney disease. We automated tubular morphometric analysis and applied it to diabetic nephropathy (DN) and transplant biopsies.
Methods
Tubules (n = 302696) were segmented with a convolutional panoptic network (Fig. 1) from 57 native DN and 30 transplant surveillance renal biopsies. Distributions of digitally quantitated tubular diameter and basement membrane (TBM) were evaluated with respect to chronic kidney disease (CKD) stage and interstitial fibrosis and tubular atrophy (IFTA) severity.
Results
The trends in Fig. 2 show that generally, as CKD and IFTA increase, tubular diameter decreases and TBM average width increases, with the DN trend being more prominent. However, significant distribution heterogeneity is observed.
Conclusion
High-throughput computation can be leveraged to automate morphometric analysis of tubules. Further data mining using a similar approach may reveal novel features that may have diagnostic or prognostic benefit.
Funding
- NIDDK Support