Scholarlink

<p>Diabetes is a major chronic disease categorized by elevating the range of sugar levels in blood cells, and early detection is crucial in reducing the risks and severity associated with the condition. Various machine learning (ML) schemes are developed to forecast the likelihood of an individual having diabetes. Classical systems lacked a Clinical Decision Support System (CDSS) using Deep Learning (DL) to predict varying diabetes levels and support clinical decisions. Therefore, a novel hybrid optimization technique Fossa Green Anaconda Optimization_Deep Kronecker Network (FGAO_DKN) is proposed for the efficient prediction of diabetes. FGAO_DKN is the amalgamation of Fossa Optimization Algorithm (FOA) and Green Anaconda Optimization (GAO) for improved feature selection and model training. The data are initially transformed using the Yeo–Johnson method. Afterward, feature selection is conducted by means of Fossa Green Anaconda Optimization (FGAO). Thereafter, data augmentation is performed with a Bootstrapping scheme. Finally, diabetes prediction is attained based on the Deep Kronecker Network (DKN). Moreover, DKN is trained with FGAO. The findings from the evaluation demonstrate that the FGAO_DKN attained an increased level of accuracy, sensitivity and specificity as 98.29%, 96.39%, and 98.59%, respectively. Thus, FGAO_DKN method offers a promising tool for healthcare professionals to make better decisions, certainly reducing the risks and severity of diabetes through timely intervention.</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

Prediction of Diabetes Using Smart Decision Support System Based on Hybrid Fossa Green Anaconda Enabled Deep Learning Approach

  • 期刊论文

  • 发表日期：

    2025年7月29日

查看全文
  • Vaibhav B. Aundhakar,
  • Mahip M. Bartere

摘要

Diabetes is a major chronic disease categorized by elevating the range of sugar levels in blood cells, and early detection is crucial in reducing the risks and severity associated with the condition. Various machine learning (ML) schemes are developed to forecast the likelihood of an individual having diabetes. Classical systems lacked a Clinical Decision Support System (CDSS) using Deep Learning (DL) to predict varying diabetes levels and support clinical decisions. Therefore, a novel hybrid optimization technique Fossa Green Anaconda Optimization_Deep Kronecker Network (FGAO_DKN) is proposed for the efficient prediction of diabetes. FGAO_DKN is the amalgamation of Fossa Optimization Algorithm (FOA) and Green Anaconda Optimization (GAO) for improved feature selection and model training. The data are initially transformed using the Yeo–Johnson method. Afterward, feature selection is conducted by means of Fossa Green Anaconda Optimization (FGAO). Thereafter, data augmentation is performed with a Bootstrapping scheme. Finally, diabetes prediction is attained based on the Deep Kronecker Network (DKN). Moreover, DKN is trained with FGAO. The findings from the evaluation demonstrate that the FGAO_DKN attained an increased level of accuracy, sensitivity and specificity as 98.29%, 96.39%, and 98.59%, respectively. Thus, FGAO_DKN method offers a promising tool for healthcare professionals to make better decisions, certainly reducing the risks and severity of diabetes through timely intervention.