Effective Processing and Analysis of Radiotherapy Images

a-Si Electronic Portal Imaging Device (EPID) is an important tool to verify the location of the radiation therapy beam with respect to the patient anatomy. But, Electronic Portal Images (EPI) suffer from low contrast. In order to have better in-treatment images to extract relevant features of the anatomy, image processing tools need to be integrated in the Radiology systems. The goal of this research work is to inspect several image processing techniques for contrast enhancement of electronic portal images and gauge parameters like mean, variance, standard deviation, MSE, RMSE, entropy, PSNR, AMBE, normalised cross correlation, average difference, structural content (SC), maximum difference and normalised absolute error (NAE) to study their visual quality improvement. In addition, by adding salt and pepper noise, Gaussian noise and motion blur, we calculate error measurement parameters like Universal Image Quality (UIQ) index, Enhancement Measurement Error (EME), Pearson Correlation Coefficient, SNR and Mean Absolute error (MAE). The improved results point out that image processing tools need to be incorporated into radiology for accurate delivery of dose.