A comparative study on feature selection for a risk prediction model for colorectal cancer

2019 ◽  
Vol 177 ◽  
pp. 219-229 ◽  
Author(s):  
Nahúm Cueto-López ◽  
Maria Teresa García-Ordás ◽  
Verónica Dávila-Batista ◽  
Víctor Moreno ◽  
Nuria Aragonés ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Feature Selection ◽  
Comparative Study ◽  
Prediction Model ◽  
Risk Prediction ◽  
Risk Prediction Model ◽  
Selection For

scholarly journals The Impact of a Comprehensive Risk Prediction Model for Colorectal Cancer on a Population Screening Program

2020 ◽  
Vol 4 (5) ◽  
Author(s):  
Sibel Saya ◽  
Jon D Emery ◽  
James G Dowty ◽  
Jennifer G McIntosh ◽  
Ingrid M Winship ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Family History ◽  
Prediction Model ◽  
Risk Prediction ◽  
Risk Prediction Model ◽  
Screening Colonoscopy ◽  
Crc Screening ◽  
Target Screening ◽  
The Impact ◽  
Risk Categories

Abstract Background In many countries, population colorectal cancer (CRC) screening is based on age and family history, though more precise risk prediction could better target screening. We examined the impact of a CRC risk prediction model (incorporating age, sex, lifestyle, genomic, and family history factors) to target screening under several feasible screening scenarios. Methods We estimated the model’s predicted CRC risk distribution in the Australian population. Predicted CRC risks were categorized into screening recommendations under 3 proposed scenarios to compare with current recommendations: 1) highly tailored, 2) 3 risk categories, and 3) 4 sex-specific risk categories. Under each scenario, for 35- to 74-year-olds, we calculated the number of CRC screens by immunochemical fecal occult blood testing (iFOBT) and colonoscopy and the proportion of predicted CRCs over 10 years in each screening group. Results Currently, 1.1% of 35- to 74-year-olds are recommended screening colonoscopy and 56.2% iFOBT, and 5.7% and 83.2% of CRCs over 10 years were predicted to occur in these groups, respectively. For the scenarios, 1) colonoscopy was recommended to 8.1% and iFOBT to 37.5%, with 36.1% and 50.1% of CRCs in each group; 2) colonoscopy was recommended to 2.4% and iFOBT to 56.0%, with 13.2% and 76.9% of cancers in each group; and 3) colonoscopy was recommended to 5.0% and iFOBT to 54.2%, with 24.5% and 66.5% of cancers in each group. Conclusions A highly tailored CRC screening scenario results in many fewer screens but more cancers in those unscreened. Category-based scenarios may provide a good balance between number of screens and cancers detected and are simpler to implement.

scholarly journals Risk Prediction Model for Colorectal Cancer: National Health Insurance Corporation Study, Korea

PLoS ONE ◽  
2014 ◽  
Vol 9 (2) ◽  
pp. e88079 ◽  
Author(s):  
Aesun Shin ◽  
Jungnam Joo ◽  
Hye-Ryung Yang ◽  
Jeongin Bak ◽  
Yunjin Park ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Health Insurance ◽  
Prediction Model ◽  
Risk Prediction ◽  
National Health Insurance ◽  
National Health ◽  
Risk Prediction Model

scholarly journals Inclusion of a Genetic Risk Score into a Validated Risk Prediction Model for Colorectal Cancer in Japanese Men Improves Performance

2017 ◽  
Vol 10 (9) ◽  
pp. 535-541 ◽  
Author(s):  
Motoki Iwasaki ◽  
Sachiko Tanaka-Mizuno ◽  
Aya Kuchiba ◽  
Taiki Yamaji ◽  
Norie Sawada ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Prediction Model ◽  
Risk Prediction ◽  
Risk Score ◽  
Genetic Risk ◽  
Genetic Risk Score ◽  
Risk Prediction Model ◽  
Japanese Men

scholarly journals Australasian ACPGBI risk prediction model for 30‐day mortality after colorectal cancer surgery

BJS Open ◽  
2020 ◽  
Vol 4 (6) ◽  
pp. 1208-1216
Author(s):  
S. Wilkins ◽  
K. Oliva ◽  
E. Chowdhury ◽  
B. Ruggiero ◽  
A. Bennett ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Prediction Model ◽  
Risk Prediction ◽  
Cancer Surgery ◽  
Risk Prediction Model ◽  
Colorectal Cancer Surgery

scholarly journals A new colorectal cancer risk prediction model incorporating family history, personal and environmental factors

2019 ◽  
Author(s):  
Yingye Zheng ◽  
Xinwei Hua ◽  
Aung K. Win ◽  
Robert J. MacInnis ◽  
Steven Gallinger ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Risk Factors ◽  
Family History ◽  
Prediction Model ◽  
Risk Prediction ◽  
Prediction Models ◽  
Familial Risk ◽  
Population Based ◽  
Risk Prediction Model ◽  
Incidence And Mortality

AbstractPurposeReducing colorectal cancer (CRC) incidence and mortality through early detection would improve efficacy if targeted. A CRC risk-prediction model incorporating personal, family, genetic and environmental risk factors could enhance prediction.MethodsWe developed risk-prediction models using population-based CRC cases (N=4,445) and controls (N=3,967) recruited by the Colon Cancer Family Registry Cohort (CCFRC). A familial risk profile (FRP) was calculated to summarize individuals’ risk based on their CRC family history, family structure, germline mutation probability in major susceptibility genes, and a polygenic component. Using logistic regression, we developed risk models including individuals’ FRP or a binary CRC family-history (FH), and risk factors collected at recruitment. Model validation used follow-up data for population-(N=12,052) and clinic-based (N=5,584) relatives with no cancer history at recruitment, assessing calibration (E/O) and discrimination (AUC).ResultsThe E/O (95% confidence interval [CI]) for FRP models for population-based relatives were 1.04 (0.74-1.45) and 0.86 (0.64-1.20) for men and women, and for clinic-based relatives 1.15 (0.87-1.58) and 1.04 (0.76-1.45). The age-adjusted AUC (95% CI) for FRP models in population-based relatives were 0.69 (0.60-0.78) and 0.70 (0.62-0.77), and for clinic-based relatives 0.77 (0.69-0.84) and 0.68 (0.60-0.76). The incremental values of AUC (95% CI) for FRP over FH models for population-based relatives were 0.08 (0.01-0.15) and 0.10 (0.04-0.16), and for clinic-based relatives 0.11 (0.05-0.17) and 0.11 (0.06-0.17).ConclusionThe FRP-based model and FH-based model calibrate well in both settings. The FRP-based model provided better risk-prediction and discrimination than the FH-based model. A detailed family history may be useful for targeted risk-based screening and clinical management.

scholarly journals Risk-adjusted colorectal cancer screening using the FIT and routine screening data: development of a risk prediction model

2017 ◽  
Vol 118 (2) ◽  
pp. 285-293 ◽  
Author(s):  
Jennifer Anne Cooper ◽  
Nick Parsons ◽  
Chris Stinton ◽  
Christopher Mathews ◽  
Steve Smith ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Cancer Screening ◽  
Prediction Model ◽  
Colorectal Cancer Screening ◽  
Risk Prediction ◽  
Risk Prediction Model ◽  
Routine Screening
Sign in / Sign up

Export Citation Format

Share Document