A Review on Detection and Classification of Osteoarthritis (OA) in Knee Radiography Images Using Deep Learning Methods
M. Elakkiya1*, Dr. R. Nithya2
Abstract
Deep learning approach is now being studied for osteoarthritis (OA) analysis in radiographic comparisons of our knee. Osteoarthritis is by far prevalent the arthritis in knee, particularly in elder persons. The degenerative joint condition osteoarthritis can lead to cartilage breakdown in our joints, that results in pain, swelling, and decreased stiffness. Due to cartilage degradation and restricted strength in osteoarthritis, the bones rub against one another and cause severe knee discomfort. In the early stages of OA, the cartilage is not severely damaged. Therefore, an automated computer-aided methodology is crucial for the early detection of osteoarthritis. Almost millions of people worldwide suffer with osteoarthritis. The current study applied and evaluated the effectiveness of deep learning techniques created to aid radiologists in early-stage knee osteoarthritis perception and classification and osteoarthritis patients. The development of orthopedic and radiology whish distresses the interpretation of radiography. In this paper the multi resolution filtering method is used to make the non-local to highlight and it isolates the sound from image. Eventually, a deep learning approach and an adaptive multi-resolution nonlocal filter are applied for this challenge. Suggested a deep learning approach based on seven methodologies: data preparation choose a model, Model training, testing, evaluation, improvement, and playback. In osteoarthritis, our planned deep learning model will more accurately identify the risk of knee replacement.
Keywords:
Artificial intelligence, Knee Osteoarthritis, Radiography images, Deep learning methods, Adaptive multi-resolution non-local filter
