Human lung cancer classification and comprehensive analysis using different machine learning techniques.

Area of Science:

• Medical Imaging
• Machine Learning
• Oncology

Background:

• Lung cancer is a leading cause of cancer-related mortality worldwide.
• Medical imaging techniques like MRI, CT, and X-ray are crucial for lung cancer diagnosis.
• Automated classification of lung cancer from imaging data is challenging due to complex image processing steps.

Purpose of the Study:

• To propose and evaluate machine learning techniques for automated human lung cancer classification.
• To compare the performance of seven different machine learning classifiers for lung cancer diagnosis.

Main Methods:

• A lung cancer dataset was utilized for image acquisition.
• Image processing techniques were applied to the input lung images.
• Seven classifiers, including k-nearest neighbors (KNN), support vector machine (SVM), decision tree (DT), multinomial naive Bayes (MNB), stochastic gradient descent (SGD), random forest (RF), and multi-layer perceptron (MLP), were employed for classification.

Main Results:

• The performance of each classifier was evaluated using metrics such as accuracy, positive predictive value, sensitivity, and f-score.
• The multi-layer perceptron (MLP) classifier achieved superior accuracy compared to other tested classifiers.
• MLP demonstrated significantly higher accuracy rates over KNN, SVM, DT, MNB, SGD, and RF.

Conclusions:

• Machine learning, particularly the MLP classifier, shows significant promise for accurate automated lung cancer classification.
• The proposed approach offers a potential advancement in the diagnostic process for lung cancer.
• Further research can build upon these findings to refine automated lung cancer detection systems.