Work

AI-Based Early Lung Cancer Detection from CT Images

Year: 2026
Type: article
Abstract: Lung cancer is one of the deadliest forms of cancer worldwide, making early detection critical for saving lives. However, reviewing large volumes of chest CT scans by hand is exhausting and time-consuming for radiologists, which can lead to missed early-stage nodules, false alarms, or differing interpretations between doctors. To help solve this, our study introduces an automated Computer-Aided Detection (CAD) system designed to identify and assess lung nodules early from CT images. The system automatically loads the scans, cleans up image noise, isolates the lung regions, and uses Convolutional Neural Networks (CNN/3D-CNN) to spot suspicious nodules and calculate their likelihood of being cancerous. Alongside image analysis, a clinical decision tree estimates the cancer stage by factoring in tumor size, biomarker levels, and biopsy results. Because doctors need to trust what an AI flags, we integrated Grad-CAM heatmaps and bounding boxes to visually show why a region was marked. Combined with growth tracking over time and automated diagnostic reporting, this system aims to reduce radiologist workload and make early lung cancer screening faster, clearer, and more dependable.
Language: English
Open Access status: green
OpenAlex ID: W7236499951
FWCI:
Cites:
Cited by:
Related to: 10