CT colonography (CTC) is a minimally invasive method for detection of colorectal polyps and colon cancer. Limitations of CTC related to the efficiency as well as the sensitivity of radiologists. Additionally, the patient's preparation was considered burdensome and the X-ray radiation that is inherent to the technique increases the risk of cancer induction. In this thesis, computerized techniques from the fields of image processing and pattern recognition are proposed in order to increase the efficiency and the acceptance of CT colonography. An automated polyp detection method is introduced to assist the radiologist. This computer aided detection (CAD) system can be used as a second reader, since it is highly sensitive and may therefore enhance the observer's sensitivity. Additionally, the reading time may be reduced as only a few false positives are detected. The CAD system finds and classifies polyp candidates based on a measure indicating the amount of protrusion into the colon. Such protruding candidates are found by using a second principal curvature flow algorithm, which makes use of the knowledge about the normal colon shape. For classification of the candidates, a low-complex pattern recognition system is designed which is shown to be highly sensitive as well as robust to data from different medical centers. Furthermore, an extended electronic cleansing algorithm is proposed that facilitates 3D reading of data from patients adhering to a limited bowel preparation. The electronic cleansing algorithm relies on a preprocessing step using the same principal curvature flow technique that was previously introduced for automated polyp detection. As such, data from patients with a limited bowel preparation can be assessed with an unfolded cube fly-through visualization method, while it does not degradate the radiologist's detection performance. Lastly, the effect of reduced radiation dose is investigated. Therefore, a technique is developed for simulating low-dose Computed Tomography (CT) scans from reconstructed high-dose images. Essentially, this enables in-silico studies into the minimal dose for a particular diagnostic task. It is used to investigate the effectiveness of the automated polyp detection system when the radiation dose is minimized. In conclusion, this thesis presents novel techniques and results that open the way to large-scale screening of colorectal polyps and colon cancer using CT colonography.