CNN-Based Lung Cancer Detection from CT Scan and X-Ray Images: A Deep Learning Perspective
International Journal of Advanced Research in Science Communication and Technology · 23.03.2026 · 393
DOI: 10.48175/ijarsct-31954 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 23.03.2026. Verfasst von Omkar A. Gaikwad und 3 weiteren Autor:innen. Die Arbeit nennt 12 Literaturangaben.
Abstract (Originalsprache)
Lung cancer remains the leading cause of cancer-related deaths globally, accounting for approximately 2.2 million new cases per year (WHO, 2024). Early and accurate detection is critical — the five-year survival rate for localised lung cancer is 59%, compared to only 6% when diagnosed at a distant stage. This paper presents an intelligent, web-based Lung Cancer Detection System that leverages Transfer Learning on the MobileNetV2 Convolutional Neural Network (CNN) architecture to analyse both Chest X-ray images and CT scan images with high accuracy. The system accepts flexible input — either X-ray alone, CT scan alone, or both simultaneously — with a weighted confidence fusion strategy that improves overall diagnostic reliability. The backend is powered by Flask (Python), which handles image pre-processing (resizing to 224 × 224 pixels, RGB normalisation), model inference, and weighted confidence calculation. Two independently trained MobileNetV2 models — one for X-ray images (87.5% accuracy) and one for CT scans (91.2% accuracy) — generate probability scores that are combined with patient risk factors (age, smoking history, family history) to produce a final diagnosis with a risk level classification (High/Moderate/Low). A key innovation of this project is the integration of a custom Blockchain module that immutably records every diagnosis. Using SHA-256 cryptographic hashing and Proof-of-Work (PoW) mining with difficulty level 3, each diagnosis is stored as a block in a verifiable chain, providing a tamper-proof audit trail for medical records. The system achieves an end-to-end response time of 2–5 seconds
Zitieren
APA
Gaikwad, O. A., Nagaonkar, J. S., Deshmukh, S. R., & Gharat, R. (2026). CNN-Based Lung Cancer Detection from CT Scan and X-Ray Images: A Deep Learning Perspective. International Journal of Advanced Research in Science Communication and Technology, 393. https://doi.org/10.48175/ijarsct-31954
Vancouver
Gaikwad OA, Nagaonkar JS, Deshmukh SR, Gharat R. CNN-Based Lung Cancer Detection from CT Scan and X-Ray Images: A Deep Learning Perspective. International Journal of Advanced Research in Science Communication and Technology. 2026:393. doi:10.48175/ijarsct-31954
BibTeX
@article{Gaikwad2026,
author = {Gaikwad, Omkar A. and Nagaonkar, Juhi S. and Deshmukh, Siddhi R. and Gharat, Reena},
title = {CNN-Based Lung Cancer Detection from CT Scan and X-Ray Images: A Deep Learning Perspective},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {393},
doi = {10.48175/ijarsct-31954}
}
EUVIMED