This work proposes a new restoration method to improve mammographic images by using Anscombe transform and Wiener filter to quantum noise reduction. Besides, it is performed an image enhancement by using a restoration inverse filter, calculated based on the image system modulation transfer function (MTF). This pre-processing technique were used for a set of mammographic phantom images in order measure the number of micro calcifications correctly detected by a computer-aided detection (CAD) algorithm. Results showed that the proposed method improved breast images quality by overcome the acquisition process constrains and reducing noise.The performance of the breast microcalcification CAD was improved when using the restored images set in comparison to the original one.