Assessment of digital x-ray machine image quality using quantitative analysis

Abstract

Introduction: The physicians and radiography specialist are concern to image quality that reveals the pathology and hence the proper management of the patient condition and be satisfied; this situation mostly lead to high dose given to the patient where if the image quality was not suitable so repetition of the imaging process is mandatory. This study includes the significant increase of image quality and the new potential for dose reduction. It intends to assess digital x-ray machine image quality using quantitative analysis for five organs; Feet exam, breast exam, Pelvic exam, Spinal exam and chest exam in police Hospital in Sudan in Khartoum State. Statement of the problem: The evaluation of x-ray using quality mostly done subjectively, using visual perception which gives unreliable result these true of quantitative methods were applied more accurate results can be obtained and hence correlation of problem will be most accurate. Purpose: this study aimed to assessment of digital x-ray machine image quality using quantitative analysis. Material and method: A total of 100 patients from both male and female were randomly selected from whom exposed their Feet, breast, pelvic, spinal or chest to digital X-ray machine, from Neusoft in Police Hospital in Khartoum State. Region of interest (ROI) were selected from the image using 3×3 pixels in the high intensity and low intensity region on the same image then data were extracted from these region as signal, noise, signal to noise ratio, contrast before and after image enhancement using histogram equalization function; where distribution of image intensity histogram were redistributed for better image quality in respect to visual perception. In X- ray imaging the exposure parameters used are selected according to patient weight and organ size. The Standard (FFD) of 100 cm was used for all routine examination and the chest X- rays FFD of 180 cm are used for geometrical reason. The Interactive Data Language (IDL), Statistical Package for the Social Sciences(SPSS) and Microsoft Excel programs was used. Results: signal in high intensity region before and after enhancement, was 1409.66±532.08 and 1859.09±614.52; at p = 0.05 using t-test where p was <0.0001 and t = 19.2. While noise before and after enhancement did not show an increase in the high intensity area but barley it deceases it was 43.94±104.53 and 42.08±9.42 at p =0.05 where p = 0.8 and t = 0.254 also Signal to noise ratio showed and improvement before enhancement since noises were not increases and the signal arbitrary were increased after enhancement as follows: 36.99±10.30 and 42.08±9.42. linear increased by 0.95 units per each units before the enhancement starting at 7.2 This increased were significance using t-test with t = 9.717 and p <0.0001. The values of the signal and noise in low intensity areas before enhancement were 677.83±517.26 and, 41.01±142.41while after enhancement the signal and noise were 1104.65±71345 and 24.51±8.84; it increased by 0.99 unit per each unit before the enhancement starting at 6.7 units at p <0.0001 and t = 13.057 Noise also follows the same pattern where it shown direct linear relationship it increased by 0.97 units per each units before enhancement starting at 6.9 at t = 1.652 and p =0.1. Conclusion: It concluded that the quantitative analysis is a valuable tool for digital X- Ray image estimation.