Knee Osteoarthritis Detection and Classification Using X-Rays
Knee Osteoarthritis (OA) is a prevalent degenerative joint disease that affects millions of people worldwide, causing pain and disability. Early and accurate detection and classification of Knee OA are essential for effective treatment and management. In this project, we present a comprehensive approach for Knee Osteoarthritis Detection and Classification using X-rays, leveraging state-of-the-art techniques and tools.
The project is implemented in MATLAB, a powerful environment for image processing and machine learning tasks. For the detection of Knee Osteoarthritis, we employ the Faster R-CNN (Region-based Convolutional Neural Network) algorithm, which excels in object detection tasks. This component identifies regions of interest in X-ray images that may exhibit signs of Knee OA.
Subsequently, for the classification task, we employ GoogleNet, a deep convolutional neural network renowned for its accuracy and efficiency. The GoogleNet model is trained on a dataset comprising X-ray images, classifying Knee Osteoarthritis into four distinct grades: Grade 1, Grade 2, Grade 3, and Grade 4. This classification is crucial for determining the severity of the disease, aiding in treatment decisions.
Our proposed system demonstrates remarkable performance, achieving an impressive accuracy of 96.95%. To assess its effectiveness comprehensively, we utilize a range of performance metrics, including error, precision, recall, specificity, F1_score, and Matthews Correlation Coefficient (MCC). These metrics collectively provide a comprehensive evaluation of the model’s ability to detect and classify Knee OA accurately.
In conclusion, our project addresses the critical need for precise and efficient Knee Osteoarthritis detection and classification using X-ray images. The combination of Faster R-CNN for detection and GoogleNet for classification results in a robust and accurate system that can aid healthcare professionals in diagnosing and treating Knee Osteoarthritis effectively. The achieved accuracy of 96.95% underscores the potential of this approach to enhance patient care and outcomes in the field of orthopedic medicine.
PROJECT OUTPUT VIDEO:
ALGORITHM / MODEL USED:
Faster R-CNN & GoogleNet.
- System : Pentium i3 Processor.
- Hard Disk : 500 GB.
- Monitor : 15’’ LED.
- Input Devices : Keyboard, Mouse.
- Ram : 8 GB.
- Operating system : Windows 10 Pro.
- Coding Language : MATLAB
- Tool : MATLABR2021A
Tayyaba Tariq; Zobia Suhail; Zubair Nawaz, “Knee Osteoarthritis Detection and Classification Using X-Rays”, IEEE Access (Volume: 11), 2023.