Osteoarthritis (OA) is a bone disease that mainly affects the cartilage. Even though there are many diseases that are commonly noticed in bones, one of the most dangerous diseases is OA. OA occurs due to the destruction of the cartilage bone. According to the survey given by the National Institute on Aging, it is revealed that most of the people in their old age are at the very advanced stage of OA. X-ray is the common imaging modality for analysing the severity of Osteoarthritis. When needed for advanced level of investigation, MRI scans and thermal images are also initialized. There are numerous methods for the analysis of OA from different modalities in the very early stage. These methods may be semi-automatic and automatic. But all the developed algorithms gave results based on the space width, and texture feature only and didn't provide any quantitative analysis based on any standard parameters. The main aim of this work is to present major research challenges in different OA detection methods, discuss different machine learning-based OA detection methods and analyse their performance. The research gap in the existing methods such as an empirical model for the detection of OA and the standard parameters for the measurement of bone marrow is discussed in the proposed paper.
Keyphrases
- knee osteoarthritis
- machine learning
- deep learning
- early stage
- bone marrow
- loop mediated isothermal amplification
- high resolution
- rheumatoid arthritis
- artificial intelligence
- real time pcr
- big data
- magnetic resonance imaging
- mesenchymal stem cells
- quality improvement
- squamous cell carcinoma
- convolutional neural network
- neural network
- extracellular matrix
- fluorescence imaging