International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 03 | Mar 2025
p-ISSN: 239-0072
www.irjet.net
Pancreatic Cancer Detection From CT Scan Images Using Deep Learning Akshat Jaiswal*1, Pratik Sao*2, Suman Yadav*3, Vikas Bharti Pandey*4, Anuyoksha Singh*5 *1,2,3,4 B.Tech Student, Department of Computer Science and Engineering, LCIT Bilaspur (C.G.)
*5 Assistant Professor, Department of Computer Science and Engineering, LCIT Bilaspur (C.G.)
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Pancreatic cancer is one of the most lethal
assesses the severity of the cancer and predicts the patient's future risk based on various health parameters.
cancers due to its late diagnosis and aggressive nature. Early detection significantly improves survival rates, but conventional diagnostic methods are often time-consuming and expensive. This project leverages deep learning techniques to develop an AI-based system capable of detecting pancreatic cancer from CT images. The model classifies images into normal, benign, or malignant cases and also predicts the stage of cancer. Additionally, the system estimates the future risk of developing pancreatic cancer based on patient details. A userfriendly interface is developed using Streamlit, allowing easy interaction for medical professionals. This AI-driven approach aims to provide rapid, cost-effective, and accurate pancreatic cancer detection
Beyond classification, this project introduces a predictive component that evaluates an individual’s likelihood of developing pancreatic cancer in the future. By analyzing key health indicators such as age, smoking history, family history of cancer, and Body Mass Index (BMI), a machine learningbased risk assessment model is incorporated. This feature empowers individuals with proactive health insights, allowing for early medical intervention and lifestyle modifications. To ensure accessibility and ease of use, the system is designed with a multi-step user interface developed using Streamlit, a lightweight web framework for interactive applications. The interface follows a structured workflow, including secure login authentication, patient detail entry, image upload, real-time processing, result visualization, and precautionary guidance. This step-bystep process ensures an efficient, user-friendly experience for both medical professionals and general users.
Key Words: Deep Learning, Pancreatic Cancer Detection, Medical Image Analysis, Convolutional Neural Networks (CNN), Computer-Aided Diagnosis (CAD), CT Scan Classification, Tumor Identification, Artificial Intelligence (AI) in Healthcare
1.INTRODUCTION
The integration of deep learning, machine learning, and an intuitive interface makes this project a powerful tool in assisting radiologists, reducing diagnostic workload, and improving early detection rates. By automating the analysis of CT scans and providing predictive risk assessment, this system has the potential to enhance cancer diagnostics, improve patient outcomes, and contribute to the ongoing advancements in AI-driven healthcare solutions.
Pancreatic cancer is one of the most aggressive and deadly forms of cancer, contributing significantly to global cancerrelated mortality. The primary challenge associated with pancreatic cancer is its late-stage detection, as symptoms often remain unnoticed until the disease has advanced. This late diagnosis drastically reduces treatment options and survival rates, making early detection crucial for improving patient outcomes. Current diagnostic techniques, such as biopsy, MRI, and CT scans, rely heavily on expert radiologists for interpretation. However, these traditional methods may not always provide early-stage detection, especially in resource-limited settings where access to specialized medical professionals is restricted.
2. LITERATURE REVIEW 2.1 Existing Cancer Detection Methods Pancreatic cancer is typically diagnosed using traditional medical imaging techniques and laboratory tests. One of the most common approaches is CT scan analysis, where radiologists examine cross-sectional images of the pancreas to detect abnormalities. While effective, this method is highly dependent on the expertise of radiologists, making it both time-consuming and susceptible to human error, especially in cases where early-stage tumors are subtle and difficult to identify.
In recent years, artificial intelligence (AI) and deep learning have revolutionized medical imaging by enabling automated, accurate, and efficient disease diagnosis. Deep learning models, particularly Convolutional Neural Networks (CNNs), have demonstrated exceptional performance in medical image analysis, making them highly suitable for detecting and classifying cancerous tissues. By leveraging CNN-based architectures, this project aims to develop an AIpowered system for early pancreatic cancer detection using CT scan images. The system not only classifies cases into Normal, Benign, and Malignant categories but also
© 2025, IRJET
|
Impact Factor value: 8.315
In addition to CT scans, MRI (Magnetic Resonance Imaging) and ultrasound are also widely used for pancreatic cancer detection. MRI provides detailed images of
|
ISO 9001:2008 Certified Journal
|
Page 240