ASN Kidney Week Abstract
Our CEO, Mary Dixon, and Director of Medical Device and Digital Innovation, Annie Lutz, co-authored an abstract titled “A Convolutional Neural Network for Large-Scale Segmentation of Kidneys in Autosomal Dominant Polycystic Kidney Disease” which was submitted and accepted to the American Society of Nephrology’s Kidney Week Conference. Read the full abstract on the ASN website, and below.
A Convolutional Neural Network for Large-Scale Segmentation of Kidneys in Autosomal Dominant Polycystic Kidney Disease
Luca Antiga, 1 Dr. imtiaz H. Mohiuddin, 2 Mary Dixon,3 Annie Lutz.4 1 Orobix Srl, Bergamo, Italy; 2 Otsuka Pharmaceutical, Rockville, MD; 3 Innovenn, Madison, WI; 4 Innovenn, Inc., Madison, WI.
Background
Total kidney volume (TKV), along with age and Glomerular Filtration Rate (eGFR), is an early prognostic marker of progression in autosomal dominant polycystic kidney disease (ADPKD). Current manual or semi-automated methods for estimation of TKV from imaging data are laborious, time-consuming approximations subject to human perception and experience; this has hampered a widespread adoption of TKV as a biomarker in ADPKD. We report the development and performance of a fully automated method for kidney segmentation and TKV estimation from magnetic imaging (MR) data in patients with ADPKD on a large patient cohort using a deep learning approach. In addition, we describe how such an estimate can be employed for predicting disease progression and monitoring progression, with the aim of supporting clinical management.
Methods
We employ a fully-convolutional neural network based on the volumetric U-net architecture, trained on an extensive dataset of 1620 T2-weighted magnetic resonance imaging scans extracted from the multicenter TEMPO3:4 trial (NCT00428948); expert outlines were available as ground truth. The method is validated on 490 scans, not included in the training dataset, extracted from 179 individual subjects. Based on the data from the same trial, we develop a similarity model for the prediction of the expected TKV growth over time. Results: We obtained a 90th percentile estimation error of TKV and its change over time of 13% and 11% of the baseline volume, respectively. We predict 3-year TKV based on baseline characteristics with R2 of 0.954 on the TEMPO3:4 placebo data.
Conclusions
The present work represents the first, large-scale example of fully automated TKV estimation in ADPKD that has been trained and validated on a large-scale, multi-centric dataset. When coupled with clinical data from the same trial, we demonstrate the ability of a machine learning algorithm to predict likely TKV progression. This prognostic information combined with other clinical findings may support clinical care.
Figure 1
Figure 1. Structure of the CNN employed for segmentation. Arrows are color-coded according to the operation being performed. The number of channels is denoted above each feature map (blue boxes).
Figure 2
Figure 2. Top row: scatter plot and bisector line for measured vs estimated TKV. Middle row: Bland-Altman plot for absolute measured vs estimated TKV (mean of differences; 95% confidence intervals; 5th, 10th, 90th and 95th percentiles): 11.55 ml; -104.106 ml to 127.21 ml; -50.12ml, -17.80ml, 48.81ml, 99.66ml. Bottom row: Bland-Altman plot for percentage error on TKV estimation (mean of differences; 95% confidence intervals; 5th, 10th, 90th and 95th percentiles): 1%; -47.91% to 49.92%; -24.60%, -11.45%, 14.01%, 23.25%.
Figure 3
Figure 3. Each quadrant shows MR images (top row), expert mask (middle row), predicted mask (bottom row) for three randomly selected scans in the validation set.
Acknowledgements
Serena Mognetti, Lisa Lozza, Valeria Mondiali, from Orobix and Anna McCreery from Innovenn provided continuing support for this project with funding by Otsuka.
References
[1] A. B. Chapman, O. Devuyst, K.U. Eckardt et al, Autosomal-dominant polycystic kidney disease (ADPKD): executive summary from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. International Society of Nephrology 85, 17 (2015).
[2] EMA/CHMP/SAWP/473433/2015 Product Development Scientific Support Department: Qualification Opinion: Total Kidney Volume (TKV) as a prognostic biomarker for use in clinical trials evaluating patients with Autosomal Dominant Polycystic Kidney Disease (ADPKD) 06 November 06, 2015.
[3] “Qualification of Biomarker: Total Kidney Volume in Studies for Treatment of Autosomal Dominant Polycystic Kidney Disease”, Food and Drug Administration Center for Drug Evaluation and Research Guidance for Industry, September 15, 2016.
[4] E. Higashihara, K. Nutahara, T. Okegawa et al, Kidney volume and function in autosomal dominant polycystic kidney disease. Clinical and Experimental Nephrology 18, 157 (2014).
[5] K. Bae, B. Park, H. Sun et al, Segmentation of Individual Renal Cysts from MR Images in Patients with Autosomal Dominant Polycystic Kidney Disease. Journal of the American Society of Nephrology (JASN) 8 (2013).
