Scholarlink

<p>Hepatitis is an inflammatory condition of the liver caused by viruses, alcohol, or other causes. The symptoms include jaundice and stomach discomfort. Hepatitis B and C are viral diseases that cause liver damage. Hepatitis B can be transmitted through blood, semen, and other body fluids, while hepatitis C is caused most commonly by blood. Both can lead to chronic liver disease, cirrhosis, cancer- even death left unattended. Machine Learning will, therefore, enhance hepatitis prediction through the analysis of the patient's data, such as the level of bilirubin. In this light, algorithms can forecast the onset, severity, and response to treatment in this complex liver disease, survival, and thus enable more personalized care and improvement of outcomes. The Random Forest Classification model is adopted in predicting hepatitis survival. Also, two optimization approaches are employed in enhancing Accuracy (Ac): Prairie Dog Optimization and Coati Optimization Algorithm. Random Forest with Coati Optimization reaches an Ac value of 0.991 during training. Then comes Random Forest with Prairie Dog Optimization with an Ac value of 0.972. The weakest model is Random Forest Classification with its Ac value at 0.963.</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

Predicting hepatitis survival with machine learning: a comparative analysis and model evaluation

  • 期刊论文

  • 发表日期：

    2025年7月25日

查看全文
  • Chao Meng

摘要

Hepatitis is an inflammatory condition of the liver caused by viruses, alcohol, or other causes. The symptoms include jaundice and stomach discomfort. Hepatitis B and C are viral diseases that cause liver damage. Hepatitis B can be transmitted through blood, semen, and other body fluids, while hepatitis C is caused most commonly by blood. Both can lead to chronic liver disease, cirrhosis, cancer- even death left unattended. Machine Learning will, therefore, enhance hepatitis prediction through the analysis of the patient's data, such as the level of bilirubin. In this light, algorithms can forecast the onset, severity, and response to treatment in this complex liver disease, survival, and thus enable more personalized care and improvement of outcomes. The Random Forest Classification model is adopted in predicting hepatitis survival. Also, two optimization approaches are employed in enhancing Accuracy (Ac): Prairie Dog Optimization and Coati Optimization Algorithm. Random Forest with Coati Optimization reaches an Ac value of 0.991 during training. Then comes Random Forest with Prairie Dog Optimization with an Ac value of 0.972. The weakest model is Random Forest Classification with its Ac value at 0.963.