The diagnostic images in diabetic retinopathy are fundamental for the diagnosis of this pathology. Numerous techniques of image filtering and purification have been performed to obtain the greatest amount of reliable information for the diagnosis; however, although research has contributed significantly to properly determine the reliable characteristics of the image to diagnose the pathology, there are still deficiencies in the digitization and processing of data that need to be taken into account. This paper presents the most relevant issues of some comparative studies between three digital retinal imaging techniques: a. Bayesian Approach b. Wavelet Transform Approach c. Adaptive filtering Approach Finally, it is concluded that the combination of the reviewed techniques can improve the digital processing of the diagnostic image.