Truace
Subscribe

AI chatbot was 87.0% accurate on kidney disease but found only 42.2% of early injury

Study title: Leveraging time-series electronic health records with large language models for chronic kidney disease diagnosis in primary care

Chronic kidney disease (CKD) presents a growing public health challenge in China, exacerbated by low patient awareness and limited nephrology resources. This study evaluated the potential of large language models (LLMs) to support CKD diagnosis in primary care using time-series electronic health record (EHR) data.

Read the study abstract

Chronic kidney disease (CKD) presents a growing public health challenge in China, exacerbated by low patient awareness and limited nephrology resources. This study evaluated the potential of large language models (LLMs) to support CKD diagnosis in primary care using time-series electronic health record (EHR) data. Longitudinal data were extracted from the EHR system of Weinan, China. Among 29 963 adults meeting inclusion criteria, 2300 participants were randomly sampled to generate clinical vignettes. CKD status…

By the Truvace desk1 min read1 sourceEvidence: Moderate · 63/100 — Observational study · n = 2,300 · 1 independent sourcePeer-reviewedTRV-2026-1152· cite & verify
Leveraging time-series electronic health records with large language models for chronic kidney disease diagnosis in primary care
"Bill on dialysis" by fireflythegreat is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.

In brief

Using the DeepSeek LLM with prompt engineering, we generated binary CKD diagnoses and probability scores based on EHR data. Diagnostic performance was evaluated using accuracy, sensitivity, specificity, F1 score, area under the curve (AUC) and the detection rate for early kidney injury and compared with traditional machine learning (ML) models.

Main points

  1. Chronic kidney disease (CKD) presents a growing public health challenge in China, exacerbated by low patient awareness and limited nephrology resources.
  2. This study evaluated the potential of large language models (LLMs) to support CKD diagnosis in primary care using time-series electronic health record (EHR) data.
  3. Longitudinal data were extracted from the EHR system of Weinan, China.

The gain

Using one month of health records from 2300 sampled adults in Weinan, China, an AI chatbot diagnosed chronic kidney disease with 87.0% accuracy, 89.1% sensitivity and 86.8% specificity.

The rundown

Chronic kidney disease (CKD) presents a growing public health challenge in China, exacerbated by low patient awareness and limited nephrology resources. This study evaluated the potential of large language models (LLMs) to support CKD diagnosis in primary care using time-series electronic health record (EHR) data.

Longitudinal data were extracted from the EHR system of Weinan, China. Among 29 963 adults meeting inclusion criteria, 2300 participants were randomly sampled to generate clinical vignettes.

Sources

  1. Peer-reviewedBMJ Health & Care Informatics2026-09-18

The debate