Scholarlink

<p>Diabetes Mellitus (DM) is one of the most incurable and chronic diseases that poses a great hazard to human life. A person is considered diabetic when the blood glucose level of the human body rises from the normal level due to defective insulin secretion, impaired biological effects, or both. With the advancement of living standards, diabetes has affected the daily lives of humans. Therefore, rapid and accurate diagnosis and analysis of this disease is a matter of concern. The earlier the diagnosis can be conquered, the easier it is to control. Therefore, the entire research community must develop an autonomous diagnosis system using a computational intelligence approach. Furthermore, the integrated consequences of the Internet of Things (IoT) and Artificial intelligence (AI) have led to the identification of hidden and valid&#xa0;patterns in data, from which knowledge inference systems can be developed. Many approaches based on artificial neural networks and machine learning algorithms have been developed and tested against diabetes datasets, mostly from the Pima Indian dataset. Despite significant efforts and correctly predicted accuracies of up to 99% in diagnosing diabetes, none of these approaches have reached clinical verification yet. Diabetologists and clinical experts are assumed to be neither informed correctly nor trained using computational diagnostic tools. Therefore, this survey aims to present an overview of the wide range of opportunities, up-to-date developments, latent in IoT, and ML algorithms applied to diabetes as it is viewed as an alternative diabetes diagnosis tool for clinical assessment.</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

Diabetes Monitoring and Prediction Using Computational Intelligence Techniques: A Systematic Review

  • 期刊论文

  • 发表日期：

    2025年3月10日

查看全文
  • Padmalaya Nayak,
  • J. Siva Naga Jyothi,
  • V. Harika,
  • K. Swaraja,
  • A. Sai Hanuman

摘要

Diabetes Mellitus (DM) is one of the most incurable and chronic diseases that poses a great hazard to human life. A person is considered diabetic when the blood glucose level of the human body rises from the normal level due to defective insulin secretion, impaired biological effects, or both. With the advancement of living standards, diabetes has affected the daily lives of humans. Therefore, rapid and accurate diagnosis and analysis of this disease is a matter of concern. The earlier the diagnosis can be conquered, the easier it is to control. Therefore, the entire research community must develop an autonomous diagnosis system using a computational intelligence approach. Furthermore, the integrated consequences of the Internet of Things (IoT) and Artificial intelligence (AI) have led to the identification of hidden and valid patterns in data, from which knowledge inference systems can be developed. Many approaches based on artificial neural networks and machine learning algorithms have been developed and tested against diabetes datasets, mostly from the Pima Indian dataset. Despite significant efforts and correctly predicted accuracies of up to 99% in diagnosing diabetes, none of these approaches have reached clinical verification yet. Diabetologists and clinical experts are assumed to be neither informed correctly nor trained using computational diagnostic tools. Therefore, this survey aims to present an overview of the wide range of opportunities, up-to-date developments, latent in IoT, and ML algorithms applied to diabetes as it is viewed as an alternative diabetes diagnosis tool for clinical assessment.