Scholarlink

This study develops and evaluates a predictive model for estimating glomerular filtration rate using advanced machine learning algorithms. We trained and tested seven models, including K-nearest neighbors, support vector machine, gradient boosting, AdaBoost, XGBoost, LightGBM, and linear support vector regression, with a focus on clinical parameters like creatinine, urea, sodium, potassium, age, and gender. Gradient boosting and XGBoost demonstrated superior performance, achieving over 96% accuracy and mean absolute percentage errors around 2.5%. Additionally, we reviewed recent advancements in the field and compared our findings with existing research. The comparative analysis highlights the effectiveness of our proposed models in accurately predicting laboratory test results. This underscores the potential of machine learning to enhance diagnostic accuracy and improve patient outcomes in nephrology.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Novel Systems Based on Artificial Intelligence and Numerical Algorithms for Predicting Laboratory Results: A Comparative Study of Original Automatic Prediction Model with Advances in the Field

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dawid Pawuś,
  • Tomasz Porazko,
  • Szczepan Paszkiel

摘要

This study develops and evaluates a predictive model for estimating glomerular filtration rate using advanced machine learning algorithms. We trained and tested seven models, including K-nearest neighbors, support vector machine, gradient boosting, AdaBoost, XGBoost, LightGBM, and linear support vector regression, with a focus on clinical parameters like creatinine, urea, sodium, potassium, age, and gender. Gradient boosting and XGBoost demonstrated superior performance, achieving over 96% accuracy and mean absolute percentage errors around 2.5%. Additionally, we reviewed recent advancements in the field and compared our findings with existing research. The comparative analysis highlights the effectiveness of our proposed models in accurately predicting laboratory test results. This underscores the potential of machine learning to enhance diagnostic accuracy and improve patient outcomes in nephrology.