Work in Progress — This project is actively under development as part of a term project at IIT Bhilai.
This project explores the application of Numerical Methods to analyze and process Lung X-Ray images for detecting abnormalities such as pneumonia, fibrosis, or other pulmonary conditions. By leveraging mathematical techniques like interpolation, numerical differentiation, matrix decomposition, and iterative solvers, we aim to build a computationally grounded pipeline for medical image analysis — without relying solely on black-box deep learning models.
The goal is to understand and implement the math behind the methods, making this equally a learning exercise in applied numerical computing and a practical engineering project.
- Apply numerical methods (e.g., Gaussian elimination, LU decomposition, Newton-Raphson) to image processing tasks
- Perform edge detection, noise filtering, and feature extraction using numerical differentiation and convolution
- Study pixel intensity distributions using numerical integration techniques
- Benchmark accuracy and convergence of different methods on X-Ray datasets
- Document mathematical derivations and results clearly
Lung-XRay-Numerical-Methods/
|
|-- /research # Literature reviews, reference papers, mathematical derivations
|-- /src # Source code: Python/MATLAB scripts implementing numerical methods
|-- /docs # Project documentation, reports, presentations
|-- README.md
- Language: Python 3.x
- Libraries: NumPy, SciPy, Matplotlib, OpenCV
- Dataset: NIH Chest X-Ray Dataset (or equivalent open-source dataset)
- Methods: Finite differences, interpolation, linear algebra solvers, FFT-based filtering
| Module | Status |
|---|---|
| Literature Review | In Progress |
| Dataset Collection | Pending |
| Numerical Pipeline Design | Pending |
| Implementation | Pending |
| Results & Report | Pending |
Swagat — IIT Bhilai
Mechanical Engineering | CAD & Computational Methods Enthusiast
This repository will be updated regularly as the project progresses.