Keywords:  CAD, Computer Aided Diagnosis

Computer aided diagnosis (CAD) is a relatively new technology that uses computer algorithms to analyze radiographic images and to help detect diseases. 

The CAD software process consists of the following three steps:

  1. Segmentation:  The boundary of lesions is determined and the image separated into regions inside and outside of the lesion.
  2. Image Processing (IP):  Computer algorithms are used to measure the pertinent features of the lesion.
  3. Decision-making:  Artificial Intelligence (AI) techniques are employed to make decisions whether the object, with all its features, is suggestive enough to be brought to the attention of the radiologist.

Typically a CAD system consists of three main parts: 

  1. Scanner:  Scans and digitizes the image such as for a mammogram (not needed when there is a direct digital system with a DICOM interface).
  2. Software:  Sophisticated computer programs to analyze and annotate the image and prompt the radiologist to review areas that may suggest a lesion.
  3. Viewer:  A high-resolution monitor that displays mammogram images as well as the overlay information generated by the CAD software.

CAD results are typically exchanged as a DICOM Structured Report. There are several applications for CAD, each with their own Structured Report, among them Mammography, lung nodules, and others.

Search Database