Scholarlink

<p>Early diagnosis of lung cancer greatly improves patient survival rates. This study used a large-scale dataset of 1000 patients with nine parameters. However, diagnosing lung cancer early remains a major challenge due to its impact on the human respiratory system. Currently, artificial intelligence techniques are among the most promising methods for diagnosing and detecting cancer. Various machine learning algorithms have been employed to identify lung cancer in patients. This paper evaluates the accuracy of nine classifiers: Decision Tree (DT), Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), Naïve Bayes (NB), Support Vector Machine (SVM), K Nearest Neighbor (KNN), Ensemble Learner (EL), Artificial Neural Network (ANN), and Kernel machine for detecting lung cancer at its early stages, which leads to saving lives. The study’s results demonstrated that the Decision Tree algorithm (DT) achieved the highest accuracy of 93.5% with a training time of 17.54 s, compared to the other algorithms used. This result constitutes a positive indicator for the earlier detection of this disease and provides a great opportunity to address and eliminate it. The study’s implications are significant for healthcare staff and specialists, as it will be adopted in the process of early cancer diagnosis.</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

Evaluation performance of machine-learning algorithms in diagnostic early stage lung cancer

  • 期刊论文

  • 发表日期：

    2026年5月15日

查看全文
  • Younis Kadthem Hameed,
  • Qahtan M. Yas

摘要

Early diagnosis of lung cancer greatly improves patient survival rates. This study used a large-scale dataset of 1000 patients with nine parameters. However, diagnosing lung cancer early remains a major challenge due to its impact on the human respiratory system. Currently, artificial intelligence techniques are among the most promising methods for diagnosing and detecting cancer. Various machine learning algorithms have been employed to identify lung cancer in patients. This paper evaluates the accuracy of nine classifiers: Decision Tree (DT), Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), Naïve Bayes (NB), Support Vector Machine (SVM), K Nearest Neighbor (KNN), Ensemble Learner (EL), Artificial Neural Network (ANN), and Kernel machine for detecting lung cancer at its early stages, which leads to saving lives. The study’s results demonstrated that the Decision Tree algorithm (DT) achieved the highest accuracy of 93.5% with a training time of 17.54 s, compared to the other algorithms used. This result constitutes a positive indicator for the earlier detection of this disease and provides a great opportunity to address and eliminate it. The study’s implications are significant for healthcare staff and specialists, as it will be adopted in the process of early cancer diagnosis.