Lung Cancer Detection using Python and Machine Learning

Abstract

Lung cancer is a leading cause of mortality worldwide, and improving survival rates depends heavily on early detection. Machine learning offers an effective approach for analyzing medical data to support timely diagnosis. This study evaluates the classification of lung cancer cases using medical features such as imaging results, symptoms like features, including demographic data (age, gender), environmental and lifestyle factors (air pollution, alcohol use, dust allergy, occupational hazards, genetic risk, smoking, passive smoking, balanced diet, obesity), chronic lung disease status, and various clinical symptoms (chest pain, coughing blood, fatigue, weight loss, shortness of breath, wheezing, swallowing difficulty, clubbing of finger nails, frequent cold, dry cough). All features are complete with no missing values. The models were trained and tested on a publicly available lung cancer dataset. Results indicate in diagram and various from we use the python for code and give result.

Introduction

Lung cancer remains one of the leading causes of death worldwide, and early, accurate diagnosis is critical for improving survival rates. Traditional diagnostic methods often suffer from delays and limited sensitivity, motivating the use of machine learning (ML) techniques to enhance early detection. This study explores the application of ML models to classify lung cancer risk by integrating clinical symptoms, lifestyle and environmental factors, and demographic information.

The literature review highlights the growing role of machine learning in lung cancer diagnosis. Prior studies demonstrate that ML and deep learning approaches—such as CNNs for imaging data and classifiers like SVM and Random Forest for clinical and lifestyle data—achieve higher diagnostic accuracy than traditional methods. Research consistently shows that combining clinical symptoms with risk factors such as smoking, air pollution, genetic predisposition, age, and gender improves prediction performance. Despite these advances, challenges remain in model generalizability, interpretability, and clinical integration.

The proposed methodology uses a publicly available Kaggle dataset containing 1,000 patient records with 26 attributes related to demographics, environmental exposure, lifestyle habits, genetic risk, and clinical symptoms. After data cleaning, feature selection, label encoding, and normalization, exploratory data analysis revealed strong correlations between lung cancer risk and factors such as smoking, air pollution, genetic risk, and chronic lung disease. Multiple machine learning models were trained and evaluated, including Logistic Regression, Decision Trees, Random Forest, Gradient Boosting, XGBoost, CatBoost, SVM, KNN, Naive Bayes, and a Multilayer Perceptron.

Model performance was assessed using accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrices. The results show that ensemble-based models performed best. The Random Forest classifier achieved the highest accuracy (approximately 94–97%), followed by XGBoost (92–95%) and CatBoost (90–94%). Simpler models like Logistic Regression showed lower accuracy, while SVM provided moderate performance with higher computational cost.

Overall, the study demonstrates that machine learning—particularly ensemble models—can effectively support early lung cancer risk prediction by leveraging diverse clinical, environmental, and demographic features. These approaches have strong potential to assist clinical decision-making and improve patient outcomes when further validated and integrated into healthcare practice.

Conclusion

One of the most dangerous illnesses is lung cancer, and increasing survival rates requires early detection. Lung cancer was predicted in this study by analyzing patient data using machine learning algorithms like Random Forest, K-Nearest Neighbors (KNN), and Logistic Regression. The findings demonstrate that machine learning can successfully identify lung cancer, assisting medical professionals in making quicker and more precise diagnoses. Because it could handle complicated medical data, Random Forest outperformed the other models, while Logistic Regression offered a straightforward method. All things considered, machine learning is essential for enhancing early diagnosis, decreasing human error, and supporting medical professionals\' decision-making. To further improve accuracy and dependability, future developments can concentrate on utilizing deep learning and bigger datasets.

References

[1] I. Chhillar, A. Singh Journal of The Institution of Engineers (India): Series B, 2023 - Springer \"An Insight into Machine Learning Techniques for Cancer Detection\" [2] \"Machine Learning Methods for Lung Cancer Early Detection\" International Journal of Medical Informatics, 2023, Elsevier, S. Patel, R. Kumar [3] \"Predictive Modeling for the Diagnosis of Lung Cancer Using Ensemble Methods\" Computers in Biology and Medicine, L. Zhang, M. Chen, Elsevier, 2023 [4] \"Feature Selection Methods in Machine Learning for the Identification of Lung Cancer\" Expert Systems with Applications, A. Gupta, P. Sharma, Elsevier, 2023 [5] \"Comparative Evaluation of Machine Learning Techniques for Predicting Lung Cancer\" Journal of Biomedical Informatics, 2023, Elsevier, J. Doe, M. Smith [6] \"Lung Cancer Detection Using Support Vector Machine-Based Classification\" Artificial Intelligence in Medicine, R. Brown, E. Wilson, 2023 Elsevier [7] \"Using a Random Forest Method to Determine the Stages of Lung Cancer\" BMC Medical Informatics and Decision Making, 2023-BioMed Central, K. Lee, H. Park [8] D. Martinez and S. Taylor, \"Using Decision Trees for Lung Cancer Diagnosis\" HealthInformatics Journal, 2023-SAGE Publications [9] \"Naïve Bayes Classifier in Predicting Outcomes of Lung Cancer\" M. Anderson, L. Thomas American Society of Clinical Oncology Journal of Clinical Oncology Informatics, 2023 [10] \"Detecting Lung Cancer Using the K-Nearest Neighbors Algorithm\" P. White, G. Harris Journal of Medical Systems, Springer, 2023 [11] \"Using Logistic Regression to Predict Lung Cancer\" S. Lewis and N. Walker Journal of Clinical Oncology and Cancer Research, Springer, 2023 [12] \"Hybrid Machine Learning Models for Precise Identification of Lung Cancer\" C. Allen and B. Hall IEEE 2023 Journal of Biomedical and Health Informatics

Copyright

Copyright © 2026 Snehal K. Kulkarni, Prasad T. Goyal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.