Chronic Kidney Disease Prognosis Prediction Using Transformer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Yohan, Kang, DongGyun, Park, SeHoon, Park, Sa-Yoon, Kim, Kwangsoo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917086280810496
author Lee, Yohan
Kang, DongGyun
Park, SeHoon
Park, Sa-Yoon
Kim, Kwangsoo
author_facet Lee, Yohan
Kang, DongGyun
Park, SeHoon
Park, Sa-Yoon
Kim, Kwangsoo
contents Chronic Kidney Disease (CKD) affects nearly 10\% of the global population and often progresses to end-stage renal failure. Accurate prognosis prediction is vital for timely interventions and resource optimization. We present a transformer-based framework for predicting CKD progression using multi-modal electronic health records (EHR) from the Seoul National University Hospital OMOP Common Data Model. Our approach (\textbf{ProQ-BERT}) integrates demographic, clinical, and laboratory data, employing quantization-based tokenization for continuous lab values and attention mechanisms for interpretability. The model was pretrained with masked language modeling and fine-tuned for binary classification tasks predicting progression from stage 3a to stage 5 across varying follow-up and assessment periods. Evaluated on a cohort of 91,816 patients, our model consistently outperformed CEHR-BERT, achieving ROC-AUC up to 0.995 and PR-AUC up to 0.989 for short-term prediction. These results highlight the effectiveness of transformer architectures and temporal design choices in clinical prognosis modeling, offering a promising direction for personalized CKD care.
format Preprint
id arxiv_https___arxiv_org_abs_2511_02340
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Chronic Kidney Disease Prognosis Prediction Using Transformer
Lee, Yohan
Kang, DongGyun
Park, SeHoon
Park, Sa-Yoon
Kim, Kwangsoo
Artificial Intelligence
Other Quantitative Biology
Chronic Kidney Disease (CKD) affects nearly 10\% of the global population and often progresses to end-stage renal failure. Accurate prognosis prediction is vital for timely interventions and resource optimization. We present a transformer-based framework for predicting CKD progression using multi-modal electronic health records (EHR) from the Seoul National University Hospital OMOP Common Data Model. Our approach (\textbf{ProQ-BERT}) integrates demographic, clinical, and laboratory data, employing quantization-based tokenization for continuous lab values and attention mechanisms for interpretability. The model was pretrained with masked language modeling and fine-tuned for binary classification tasks predicting progression from stage 3a to stage 5 across varying follow-up and assessment periods. Evaluated on a cohort of 91,816 patients, our model consistently outperformed CEHR-BERT, achieving ROC-AUC up to 0.995 and PR-AUC up to 0.989 for short-term prediction. These results highlight the effectiveness of transformer architectures and temporal design choices in clinical prognosis modeling, offering a promising direction for personalized CKD care.
title Chronic Kidney Disease Prognosis Prediction Using Transformer
topic Artificial Intelligence
Other Quantitative Biology
url https://arxiv.org/abs/2511.02340