Knee osteoarthritis, a degenerative joint ailment affecting a large global population, results in pain, stiffness, and diminished mobility. The severity of this condition varies among individuals, making accurate assessment crucial for effective treatment planning. Evaluation traditionally relies on observing joint space narrowing, osteophytes, bone deformity, and sclerosis in radiographic images, using the time-consuming KL, Kellgren, and Lawrence, grading system. This method demands expertise, typically from professionals with fellowship training in arthroplasty or radiography. [1].
To enhance the efficiency of KL grade evaluation, two experts independently conduct radiographic assessments, and in cases of conflicting diagnoses, discussions are held to reach a consensus. Our proposed model utilizes deep learning techniques and achieves a 95% accuracy in detecting and classifying knee osteoarthritis severity from medical images. This automated system reduces time consumption, enabling clinicians to focus on clinical findings. It serves as a potent tool, offering a precise diagnosis and suggesting a primary treatment plan for knee osteoarthritis, providing clinicians with valuable support.