Reliable and Reproducible Classification System for Scoliotic Radiograph using Image Processing Techniques

Anitha, H and Prabhu, G K and Karunakar, A K (2014) Reliable and Reproducible Classification System for Scoliotic Radiograph using Image Processing Techniques. J Med System, 38. ISSN 0148-5598

[img] PDF
p4.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

Scoliosis classification is useful for guiding the treatment and testing the clinical outcome. State-of-theart classification procedures are inherently unreliable and non-reproducible due to technical and human judgmental error. In the current diagnostic system each examiner will have diagrammatic summary of classification procedure, number of scoliosis curves, apex level, etc. It is very difficult to define the required anatomical parameters in the noisy radiographs. The classification system demands automatic image understanding system. The proposed automated classification procedures extracts the anatomical features using image processing and applies classification procedures based on computer assisted algorithms. The reliability and reproducibility of the proposed computerized image understanding system are compared with manual and computer assisted system using Kappa values.

Item Type: Article
Uncontrolled Keywords: King’s classification · Scoliosis · Inter and intra observer variation · Rule based algorithm · Gray level profiles · FFT interpolation
Subjects: Engineering > MIT Manipal > Electronics and Communication
Depositing User: MIT Library
Date Deposited: 20 Oct 2014 07:55
Last Modified: 20 Oct 2014 07:55
URI: http://eprints.manipal.edu/id/eprint/140884

Actions (login required)

View Item View Item