AI Article Synopsis

  • Diabetic nephropathy (DN) is the main reason for end-stage renal disease (ESRD) in the U.S. and presents challenges for pathologists due to its complicated spatial patterns in kidney biopsies.
  • A novel transformer-based framework is introduced that enhances ESRD prediction by employing advanced techniques like nonlinear dimensionality reduction and spatial self-attention, allowing for better contextual understanding of whole slide images (WSIs).
  • This new model significantly outperformed traditional methods, achieving a high AUC of 0.97 for predicting two-year ESRD, showcasing the potential for improved predictive accuracy in limited pathology datasets.

Article Abstract

Diabetic nephropathy (DN) in the context of type 2 diabetes is the leading cause of end-stage renal disease (ESRD) in the United States. DN is graded based on glomerular morphology and has a spatially heterogeneous presentation in kidney biopsies that complicates pathologists' predictions of disease progression. Artificial intelligence and deep learning methods for pathology have shown promise for quantitative pathological evaluation and clinical trajectory estimation; but, they often fail to capture large-scale spatial anatomy and relationships found in whole slide images (WSIs). In this study, we present a transformer-based, multi-stage ESRD prediction framework built upon nonlinear dimensionality reduction, relative Euclidean pixel distance embeddings between every pair of observable glomeruli, and a corresponding spatial self-attention mechanism for a robust contextual representation. We developed a deep transformer network for encoding WSI and predicting future ESRD using a dataset of 56 kidney biopsy WSIs from DN patients at Seoul National University Hospital. Using a leave-one-out cross-validation scheme, our modified transformer framework outperformed RNNs, XGBoost, and logistic regression baseline models, and resulted in an area under the receiver operating characteristic curve (AUC) of 0.97 (95% CI: 0.90-1.00) for predicting two-year ESRD, compared with an AUC of 0.86 (95% CI: 0.66-0.99) without our relative distance embedding, and an AUC of 0.76 (95% CI: 0.59-0.92) without a denoising autoencoder module. While the variability and generalizability induced by smaller sample sizes are challenging, our distance-based embedding approach and overfitting mitigation techniques yielded results that sugest opportunities for future spatially aware WSI research using limited pathology datasets.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9980230PMC
http://dx.doi.org/10.1101/2023.02.20.23286044DOI Listing

Publication Analysis

Top Keywords

spatially aware
8
diabetic nephropathy
8
slide images
8
aware transformer
4
transformer networks
4
networks contextual
4
contextual prediction
4
prediction diabetic
4
nephropathy progression
4
progression slide
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!