Revolutionizing Lung Cancer Detection: A High-Accuracy Machine Learning Framework for Early Diagnosis

Area of Science:

• Oncology
• Medical Informatics
• Computational Biology

Background:

• Lung cancer is a leading cause of cancer-related mortality globally, with 1.82 million deaths reported in 2024.
• Early detection of lung cancer is critical for enhancing patient survival rates and enabling timely, effective treatment strategies.
• The high disease burden necessitates advanced methods for accurate and efficient lung cancer prediction.

Purpose of the Study:

• To conduct a systematic literature review on lung cancer prediction methods.
• To develop and validate a highly accurate machine learning framework for early lung cancer detection.
• To investigate the effectiveness of artificial intelligence (AI) and machine learning (ML) in identifying lung cancer patterns and distinguishing it from patient symptoms.

Main Methods:

• A systematic literature review was performed using the Tollgate methodology and quality assessment criteria.
• Machine learning techniques were employed, including feature selection (SelectKBest) and class imbalance handling (SMOTE).
• A voting ensemble model incorporating Random Forest, Support Vector Machine, and Logistic Regression with cross-validation was developed.

Main Results:

• The proposed machine learning framework achieved high prediction accuracy: 99% on the first dataset and 92.5% on the second.
• The study identified key features distinguishing lung cancer from patient symptoms through ML analysis.
• The systematic review addressed four research questions regarding ML/AI in lung cancer prediction.

Conclusions:

• The developed machine learning framework demonstrates significant potential for accurate and early lung cancer prediction.
• AI and ML approaches show promise in outperforming traditional methods for lung cancer diagnosis.
• This research underscores the importance of advanced computational methods in improving lung cancer outcomes.