You must login before you can run this tool.
Patient-specific colorectal cancer prediction model
This model predicts the likely findings of colonoscopy exams based on a patient's demographic characteristics and endoscopic history.
Version 2.1 - published on 17 Sep 2011
This tool is closed source.
| Category | Tools |
|---|---|
| Abstract | Risk factors for colorectal cancer (CRC) differ between patients and can include characteristics such as gender, age, and family history of CRC. In addition, clinical risk factors such finding multiple or advanced adenomas at a baseline colonoscopy have shown even stronger ties to CRC risk than the above demographic characteristics. This model utilizes an individual patient’s demographic risk factors and previous endoscopic history to predict the likelihood of finding colorectal cancer or other colonic neoplasia at a colonoscopy. The mathematical model assumes that the majority of colorectal cancers develop from pre-cancerous neoplasia called adenomas. As these adenomas grow and increase in size, they may be visible during a colonoscopy exam. Besides detection, a colonoscopy allows for the physical removal of adenomas; this short-circuits the development of potential future colorectal cancer and decreases its risk. Clinical literature has shown that the presence of adenomas and certain patient characteristics are asocial with an increased risk for developing adenomas and colorectal cancer. As such, this model makes predictions based on patient characteristics and then adapts those predictions based on the results of colonoscopic exams. |
| Cite this work | Researchers should cite this work as follows: |