We propose an automated technique for detecting fractures in X-ray images. Traditionally, X-ray images are examined manually, which is time-consuming and prone to human error. Since X-ray images are often affected by noise, multiple preprocessing steps are applied to enhance image clarity by removing noise and blur. This improves the system’s ability to detect fractures more accurately.
The system identifies fractures based on their type. By transforming noisy images into clearer versions, the detection process becomes more effective. Image processing techniques are used to track bone structures, eliminating unwanted and smaller objects. Finally, the system detects fractures using a connected component analysis and highlights them with a bounding box.
The proposed system achieves the following accuracy rates:
- Bone dislocation: 80% success rate
- Major fractures: 60-70% accuracy
- Minor fractures: 50-60% accuracy
Features:
- The orthopedic surgeon provides an X-ray image of the fractured bone.
- Image processing tools remove unwanted elements from the image.
- The system detects whether a bone is fractured and estimates the number of cracks.
- This system saves time for orthopedic surgeons, allowing them to treat patients more efficiently.
Software & Hardware Requirements:
Software:
- Version: MATLAB R2013a
Hardware:
- Processor: Intel i3
- Operating System: Windows 7 and above
Advantages:
? Saves time
? Provides accurate results
Disadvantages:
? Requires high-quality images for optimal performance
Application:
- Orthopedic surgeons
