Scholarlink

Objective <p>The study systemically reviews the existing prediction model of diabetic kidney disease (DKD) using artificial intelligence (AI) based on machine learning (ML) methods to identify the vital predictive factors, explore research gaps in next-generation prediction models, and highlight the limitations of existing studies.</p> Methods <p>The systematic review performed by exploring the well-known journal database SCOPUS, EMBASE, PUBMED, and the DBLP Computer Science Bibliography, and follows Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) approach. Further the data is extracted using Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS). The Risk of Bias is accessed by PROBAST of all the included studies.</p> Results <p>Screening from 4754 related papers, 23 studies are taken into consideration. The study type, settings, publication year, sample sizes, modeling approaches, and risk factors associated with DKD are summarized. The limitations of each study are also mentioned. In addition, a list of the Risk of Bias including all the domains and the applicability concern is provided.</p> Conclusion <p>Incorporating advanced techniques in modeling and clinical applications is expected to improve the predictive accuracy and validation of studies on DKD.</p>

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

A systematic review on machine learning based predictive modeling for diabetic kidney disease (DKD)

  • 期刊论文

  • 发表日期：

    2026年4月1日

查看全文
  • Subhashree Palaur,
  • Manoj Ranjan Mishra,
  • Nikunj Kishore Rout,
  • Rekha Sahu,
  • Satya Ranjan Dash,
  • Rajani Kanta Mahapatra

摘要

Objective

The study systemically reviews the existing prediction model of diabetic kidney disease (DKD) using artificial intelligence (AI) based on machine learning (ML) methods to identify the vital predictive factors, explore research gaps in next-generation prediction models, and highlight the limitations of existing studies.

Methods

The systematic review performed by exploring the well-known journal database SCOPUS, EMBASE, PUBMED, and the DBLP Computer Science Bibliography, and follows Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) approach. Further the data is extracted using Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS). The Risk of Bias is accessed by PROBAST of all the included studies.

Results

Screening from 4754 related papers, 23 studies are taken into consideration. The study type, settings, publication year, sample sizes, modeling approaches, and risk factors associated with DKD are summarized. The limitations of each study are also mentioned. In addition, a list of the Risk of Bias including all the domains and the applicability concern is provided.

Conclusion

Incorporating advanced techniques in modeling and clinical applications is expected to improve the predictive accuracy and validation of studies on DKD.