Image processing

Computer science

Image segmentation

Radiology

Alternative medicine

Medical imaging

Segmentation

Pattern recognition (psychology)

Image (mathematics)

Artificial intelligence

Cervical spondylosis

Medicine

Magnetic resonance imaging

Computer vision

Pathology

Cervical Spondylosis (CS) is one of the most common chronic serious bone diseases that mainly affects patient's health and leads to human death also. It basically occurs in old age patients but spreading very much in young age patients too based on different age factors and side effects. Cervical Spondylosis (CS) is becoming most important research area in medical science field and medical image processing is very essential for this research study. Mainly, it causes mild pain in disk points of the neck, spinal cord and some other facts can build Cervical Spondylosis. There is high significance in image processing techniques, segmentation techniques and point-based techniques for early detection of error at particular point where the changes can be identified in few points of neck, spinal cord or other bones of human body. Identification of these changes, mild symptoms from images like X-ray, Computed Tomography (CT) scan or Magnetic Resonance Images (MRI) are very complex for further diagnosis and identification of spondylosis. So early detection of spondylosis using medical image processing techniques gives better time-consuming, costliness of hospital medical service reduces the risk level diagnosis also. In this research work, MRI cervical images are collected and analysis is carried out by using Principle Component Analysis (PCA) segmentation classifier image processing technique for early detection of spondylosis. By applying this method, performance metrics like Mean Square Error (MSE), Peak Signal Noise Ratio (PSNR) and Elapsed Time (ET) are calculated for input image −1 and achieved 0.000095 error value in particular point where the changes are identified easily with better Elapsed Time complexity 3.7 (in Seconds), but for input image – 2 achieved 0.000005 error value in particular point where the changes are identified easily with better Elapsed Time complexity 2.5 (in Seconds) less execution time.

Early Detection of Spondylosis using Point-Based Image Processing Techniques