Cargando…

Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images

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...

Descripción completa

Detalles Bibliográficos
Autores principales: Shickel, Benjamin, Lucarelli, Nicholas, Rao, Adish S., Yun, Donghwan, Moon, Kyung Chul, Han, Seung Seok, Sarder, Pinaki
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Cold Spring Harbor Laboratory 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9980230/
https://www.ncbi.nlm.nih.gov/pubmed/36865174
http://dx.doi.org/10.1101/2023.02.20.23286044
_version_ 1784899873894563840
author Shickel, Benjamin
Lucarelli, Nicholas
Rao, Adish S.
Yun, Donghwan
Moon, Kyung Chul
Han, Seung Seok
Sarder, Pinaki
author_facet Shickel, Benjamin
Lucarelli, Nicholas
Rao, Adish S.
Yun, Donghwan
Moon, Kyung Chul
Han, Seung Seok
Sarder, Pinaki
author_sort Shickel, Benjamin
collection PubMed
description 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.
format Online
Article
Text
id pubmed-9980230
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Cold Spring Harbor Laboratory
record_format MEDLINE/PubMed
spelling pubmed-99802302023-03-03 Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images Shickel, Benjamin Lucarelli, Nicholas Rao, Adish S. Yun, Donghwan Moon, Kyung Chul Han, Seung Seok Sarder, Pinaki medRxiv Article 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. Cold Spring Harbor Laboratory 2023-02-23 /pmc/articles/PMC9980230/ /pubmed/36865174 http://dx.doi.org/10.1101/2023.02.20.23286044 Text en https://creativecommons.org/licenses/by-nc-nd/4.0/This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/) , which allows reusers to copy and distribute the material in any medium or format in unadapted form only, for noncommercial purposes only, and only so long as attribution is given to the creator.
spellingShingle Article
Shickel, Benjamin
Lucarelli, Nicholas
Rao, Adish S.
Yun, Donghwan
Moon, Kyung Chul
Han, Seung Seok
Sarder, Pinaki
Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title_full Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title_fullStr Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title_full_unstemmed Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title_short Spatially Aware Transformer Networks for Contextual Prediction of Diabetic Nephropathy Progression from Whole Slide Images
title_sort spatially aware transformer networks for contextual prediction of diabetic nephropathy progression from whole slide images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9980230/
https://www.ncbi.nlm.nih.gov/pubmed/36865174
http://dx.doi.org/10.1101/2023.02.20.23286044
work_keys_str_mv AT shickelbenjamin spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT lucarellinicholas spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT raoadishs spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT yundonghwan spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT moonkyungchul spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT hanseungseok spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages
AT sarderpinaki spatiallyawaretransformernetworksforcontextualpredictionofdiabeticnephropathyprogressionfromwholeslideimages