https://github.com/akimuddinshaikh/deep-learning-projects
Bone Fracture Detection Using X-Ray Images An AI-driven approach for detecting bone fractures in X-ray images using Convolutional Neural Networks (CNNs), Canny Edge Detection, Random Forest, and Transfer Learning (EfficientNetB3). Achieved 94.59% accuracy with advanced preprocessing techniques.
https://github.com/akimuddinshaikh/deep-learning-projects
cnn-classification deep-learning medical-imaging python xray-classification-cdac
Last synced: 4 months ago
JSON representation
Bone Fracture Detection Using X-Ray Images An AI-driven approach for detecting bone fractures in X-ray images using Convolutional Neural Networks (CNNs), Canny Edge Detection, Random Forest, and Transfer Learning (EfficientNetB3). Achieved 94.59% accuracy with advanced preprocessing techniques.
- Host: GitHub
- URL: https://github.com/akimuddinshaikh/deep-learning-projects
- Owner: Akimuddinshaikh
- Created: 2025-02-04T02:02:51.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-02-04T02:07:09.000Z (4 months ago)
- Last Synced: 2025-02-04T03:18:50.389Z (4 months ago)
- Topics: cnn-classification, deep-learning, medical-imaging, python, xray-classification-cdac
- Language: Jupyter Notebook
- Homepage:
- Size: 4.36 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Bone Fracture Detection Using X-Ray Images
DMML2: Project Report
Author: Akimuddin Aslam Shaikh and Sanjay Girish Dialani
Institution: National College of Ireland, Dublin, Ireland
📌 Project Overview
This project aims to develop an automated bone fracture detection system using machine learning and deep learning techniques. It focuses on analyzing X-ray images to diagnose fractures efficiently and accurately, assisting healthcare professionals in making informed decisions.
🔍 Key Features
✅ Implemented Convolutional Neural Networks (CNNs) for deep learning-based feature extraction.
✅ Applied Canny Edge Detection combined with Random Forest for classical machine learning-based fracture detection.
✅ Used Transfer Learning with a pre-trained EfficientNetB3 model to improve accuracy.
✅ Followed the CRISP-DM methodology for data preprocessing, feature engineering, and evaluation.
✅ Achieved high accuracy (94.59%) with EfficientNetB3 and Edge Detection, improving diagnostic reliability.