LUNG CANCER PREDICTION MODEL

Abstract Lung cancer remains one of the leading causes of cancer-related mortality worldwide. Early identification of high-risk individuals can improve outcomes by enabling timely diagnostic workup and intervention. This paper presents a lightweight, end to-end system for lung cancer risk prediction based on a logistic-regression model trained on tabular lifestyle and symptom features and deployed as a Flask-based web application. Using a synthetic dataset that encodes age, smoking duration, air pollution exposure, alcohol intake, and key respiratory symptoms, we train a standardized logistic regression classifier and evaluate its performance with ROC– AUC and standard classification metrics. The trained model is exposed through both a graphical web interface and a JSON API, illustrating how traditional machine-learning models can be integrated into a modern web stack. Although the dataset is synthetic and the system is not intended for clinical use, it provides a practical template for building interpretable, low latency medical decision-support prototypes.