Click here to


Are you sure ?

Yes, do it No, cancel

Risk-Index of Colorectal Cancer to Triage for Screening

B. Nartowt*, G. Hart , D. Roffman , I. Ali , W. Muhammad , Y. Liang , J. Deng , Yale university School of Medicine, New Haven, CT


(Tuesday, 7/31/2018) 3:45 PM - 4:15 PM

Room: Exhibit Hall | Forum 9

Purpose: Colorectal cancer (CRC) is the unregulated growth of one or more adenomatous polyps occurring in the colon and/or the rectum. Of all new US cancers cases 8.0% are colorectal. CRC claims 8.4% of all cancer-deaths (an above-average mortality). For early stage detection, the United States Preventative Services Task Force (USPSTF) recommends screening for ages 50-75. However, the high false positive rate of these guidelines leads to much unnecessary, expensive, and occasionally injurious screening. Hence, an artificial neural network (ANN) for CRC risk prediction usable for triaging people for screening based on their personal health data.

Methods: The 1997-2016 responses to the National Health Interview Survey (NHIS) personal health questionnaire were used to train and validate an ANN. As a binary test, sensitivity (TPR), specificity (SPC), and positive/negative predictive values (PPV/NPV) are calculated and compared with the USPSTF. As a trinary test, three levels of risk-stratification (for optional, biennial, and annual screening) are defined by requiring that ≤1% of CRC/non-CRC cases be misclassified as low/high-risk.

Results: As a binary test, the ANN has sensitivity of ~0.7, specificity of ~0.7, PPV of ~0.09, and NPV of ~0.997, all (except for NPV) exceeding the USPSTF guidelines and independent of tumorous advancement. As a trinary test, lowered SPC and PPV are exchanged for higher TPR and NPV.

Conclusion: CRC risk calculated by ANN is noninvasive, insensitive to tumorous advancement, & outperforms USPSTF screening guidelines as a binary and trinary test. In addition, the ANN offers the prospect of CRC risk assessment in real time and on the world map.


Data Interpolation, Risk, ROC Analysis


TH- Dataset analysis/biomathematics: Machine learning techniques

Contact Email