Scholarlink

Fatty Liver Disease (FLD) is a prevalent health concern, demanding accurate predictive models for timely intervention and effective management. This paper presents a comprehensive investigation into FLD prediction employing various machine learning (ML) models, each meticulously analysed for performance metrics. The classifiers considered include Naïve Bayes (N.B), Random Forest (R.F), Logistic Regression (L.R), Convolutional Neural Network (C.N.N), XGBoost, Decision Tree, Gated Recurrent Unit (G.R.U), Extra Trees, Long Short Term Memory (L.S.T.M), K Nearest Neighbour (K.N.N), Support Vector Machine (S.V.M), and Recurrent Neural Network (R.N.N). Results showcase the efficacy of these models, revealing substantial variations in evaluation metrics. Notably, Extra Trees demonstrated exceptional performance, achieving a remarkable accuracy of 99.25%. Decision Tree also exhibited notable accuracy of 95.28%. Random Forest, C.N.N, G.R.U, and K.N.N demonstrated commendable performance, offering a diverse range of choices for FLD prediction tailored to specific clinical requirements. These findings underscore the potential of ML models in revolutionizing FLD diagnosis, facilitating personalized patient care, and optimizing healthcare resources. This research contributes valuable insights into the comparative performance of diverse ML models for FLD prediction, paving the way for future developments in the integration of these models into clinical practice. This paper holds promise for advancing predictive analytics in FLD detection and exemplify the evolving role of ML in enhancing diagnostic accuracy and patient outcomes.

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

Liver Disease Diagnosis Using Machine Learning and Deep Learning Techniques: An Experimental Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Satyam Kumar,
  • Anirudh Singh,
  • Pragya Prasad,
  • Deepjyoti Choudhury

摘要

Fatty Liver Disease (FLD) is a prevalent health concern, demanding accurate predictive models for timely intervention and effective management. This paper presents a comprehensive investigation into FLD prediction employing various machine learning (ML) models, each meticulously analysed for performance metrics. The classifiers considered include Naïve Bayes (N.B), Random Forest (R.F), Logistic Regression (L.R), Convolutional Neural Network (C.N.N), XGBoost, Decision Tree, Gated Recurrent Unit (G.R.U), Extra Trees, Long Short Term Memory (L.S.T.M), K Nearest Neighbour (K.N.N), Support Vector Machine (S.V.M), and Recurrent Neural Network (R.N.N). Results showcase the efficacy of these models, revealing substantial variations in evaluation metrics. Notably, Extra Trees demonstrated exceptional performance, achieving a remarkable accuracy of 99.25%. Decision Tree also exhibited notable accuracy of 95.28%. Random Forest, C.N.N, G.R.U, and K.N.N demonstrated commendable performance, offering a diverse range of choices for FLD prediction tailored to specific clinical requirements. These findings underscore the potential of ML models in revolutionizing FLD diagnosis, facilitating personalized patient care, and optimizing healthcare resources. This research contributes valuable insights into the comparative performance of diverse ML models for FLD prediction, paving the way for future developments in the integration of these models into clinical practice. This paper holds promise for advancing predictive analytics in FLD detection and exemplify the evolving role of ML in enhancing diagnostic accuracy and patient outcomes.