Cargando…

Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma

Screening for colorectal cancer (CRC) continues to rely on colonoscopy and/or fecal occult blood testing since other (non-invasive) risk-stratification systems have not yet been implemented into European guidelines. In this study, we evaluate the potential of machine learning (ML) methods to predict...

Descripción completa

Detalles Bibliográficos
Autores principales: Semmler, Georg, Wernly, Sarah, Wernly, Bernhard, Mamandipoor, Behrooz, Bachmayer, Sebastian, Semmler, Lorenz, Aigner, Elmar, Datz, Christian, Osmani, Venet
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8538127/
https://www.ncbi.nlm.nih.gov/pubmed/34683122
http://dx.doi.org/10.3390/jpm11100981
_version_ 1784588430752088064
author Semmler, Georg
Wernly, Sarah
Wernly, Bernhard
Mamandipoor, Behrooz
Bachmayer, Sebastian
Semmler, Lorenz
Aigner, Elmar
Datz, Christian
Osmani, Venet
author_facet Semmler, Georg
Wernly, Sarah
Wernly, Bernhard
Mamandipoor, Behrooz
Bachmayer, Sebastian
Semmler, Lorenz
Aigner, Elmar
Datz, Christian
Osmani, Venet
author_sort Semmler, Georg
collection PubMed
description Screening for colorectal cancer (CRC) continues to rely on colonoscopy and/or fecal occult blood testing since other (non-invasive) risk-stratification systems have not yet been implemented into European guidelines. In this study, we evaluate the potential of machine learning (ML) methods to predict advanced adenomas (AAs) in 5862 individuals participating in a screening program for colorectal cancer. Adenomas were diagnosed histologically with an AA being ≥ 1 cm in size or with high-grade dysplasia/villous features being present. Logistic regression (LR) and extreme gradient boosting (XGBoost) algorithms were evaluated for AA prediction. The mean age was 58.7 ± 9.7 years with 2811 males (48.0%), 1404 (24.0%) of whom suffered from obesity (BMI ≥ 30 kg/m²), 871 (14.9%) from diabetes, and 2095 (39.1%) from metabolic syndrome. An adenoma was detected in 1884 (32.1%), as well as AAs in 437 (7.5%). Modelling 36 laboratory parameters, eight clinical parameters, and data on eight food types/dietary patterns, moderate accuracy in predicting AAs with XGBoost and LR (AUC-ROC of 0.65–0.68) could be achieved. Limiting variables to established risk factors for AAs did not significantly improve performance. Moreover, subgroup analyses in subjects without genetic predispositions, in individuals aged 45–80 years, or in gender-specific analyses showed similar results. In conclusion, ML based on point-prevalence laboratory and clinical information does not accurately predict AAs.
format Online
Article
Text
id pubmed-8538127
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-85381272021-10-24 Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma Semmler, Georg Wernly, Sarah Wernly, Bernhard Mamandipoor, Behrooz Bachmayer, Sebastian Semmler, Lorenz Aigner, Elmar Datz, Christian Osmani, Venet J Pers Med Article Screening for colorectal cancer (CRC) continues to rely on colonoscopy and/or fecal occult blood testing since other (non-invasive) risk-stratification systems have not yet been implemented into European guidelines. In this study, we evaluate the potential of machine learning (ML) methods to predict advanced adenomas (AAs) in 5862 individuals participating in a screening program for colorectal cancer. Adenomas were diagnosed histologically with an AA being ≥ 1 cm in size or with high-grade dysplasia/villous features being present. Logistic regression (LR) and extreme gradient boosting (XGBoost) algorithms were evaluated for AA prediction. The mean age was 58.7 ± 9.7 years with 2811 males (48.0%), 1404 (24.0%) of whom suffered from obesity (BMI ≥ 30 kg/m²), 871 (14.9%) from diabetes, and 2095 (39.1%) from metabolic syndrome. An adenoma was detected in 1884 (32.1%), as well as AAs in 437 (7.5%). Modelling 36 laboratory parameters, eight clinical parameters, and data on eight food types/dietary patterns, moderate accuracy in predicting AAs with XGBoost and LR (AUC-ROC of 0.65–0.68) could be achieved. Limiting variables to established risk factors for AAs did not significantly improve performance. Moreover, subgroup analyses in subjects without genetic predispositions, in individuals aged 45–80 years, or in gender-specific analyses showed similar results. In conclusion, ML based on point-prevalence laboratory and clinical information does not accurately predict AAs. MDPI 2021-09-29 /pmc/articles/PMC8538127/ /pubmed/34683122 http://dx.doi.org/10.3390/jpm11100981 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Semmler, Georg
Wernly, Sarah
Wernly, Bernhard
Mamandipoor, Behrooz
Bachmayer, Sebastian
Semmler, Lorenz
Aigner, Elmar
Datz, Christian
Osmani, Venet
Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title_full Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title_fullStr Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title_full_unstemmed Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title_short Machine Learning Models Cannot Replace Screening Colonoscopy for the Prediction of Advanced Colorectal Adenoma
title_sort machine learning models cannot replace screening colonoscopy for the prediction of advanced colorectal adenoma
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8538127/
https://www.ncbi.nlm.nih.gov/pubmed/34683122
http://dx.doi.org/10.3390/jpm11100981
work_keys_str_mv AT semmlergeorg machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT wernlysarah machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT wernlybernhard machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT mamandipoorbehrooz machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT bachmayersebastian machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT semmlerlorenz machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT aignerelmar machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT datzchristian machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma
AT osmanivenet machinelearningmodelscannotreplacescreeningcolonoscopyforthepredictionofadvancedcolorectaladenoma