Cargando…
Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy
This study retrospectively analyzed the performance of artificial neural networks (ANN) to predict overall survival (OS) or locoregional failure (LRF) in HNSCC patients undergoing radiotherapy, based on 2-[(18)F]FDG PET/CT and clinical covariates. We compared predictions relying on three different s...
Autores principales: | , , , , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8468242/ https://www.ncbi.nlm.nih.gov/pubmed/34573924 http://dx.doi.org/10.3390/diagnostics11091581 |
_version_ | 1784573614738112512 |
---|---|
author | Marschner, Sebastian N. Lombardo, Elia Minibek, Lena Holzgreve, Adrien Kaiser, Lena Albert, Nathalie L. Kurz, Christopher Riboldi, Marco Späth, Richard Baumeister, Philipp Niyazi, Maximilian Belka, Claus Corradini, Stefanie Landry, Guillaume Walter, Franziska |
author_facet | Marschner, Sebastian N. Lombardo, Elia Minibek, Lena Holzgreve, Adrien Kaiser, Lena Albert, Nathalie L. Kurz, Christopher Riboldi, Marco Späth, Richard Baumeister, Philipp Niyazi, Maximilian Belka, Claus Corradini, Stefanie Landry, Guillaume Walter, Franziska |
author_sort | Marschner, Sebastian N. |
collection | PubMed |
description | This study retrospectively analyzed the performance of artificial neural networks (ANN) to predict overall survival (OS) or locoregional failure (LRF) in HNSCC patients undergoing radiotherapy, based on 2-[(18)F]FDG PET/CT and clinical covariates. We compared predictions relying on three different sets of features, extracted from 230 patients. Specifically, (i) an automated feature selection method independent of expert rating was compared with (ii) clinical variables with proven influence on OS or LRF and (iii) clinical data plus expert-selected SUV metrics. The three sets were given as input to an artificial neural network for outcome prediction, evaluated by Harrell’s concordance index (HCI) and by testing stratification capability. For OS and LRF, the best performance was achieved with expert-based PET-features (0.71 HCI) and clinical variables (0.70 HCI), respectively. For OS stratification, all three feature sets were significant, whereas for LRF only expert-based PET-features successfully classified low vs. high-risk patients. Based on 2-[(18)F]FDG PET/CT features, stratification into risk groups using ANN for OS and LRF is possible. Differences in the results for different feature sets confirm the relevance of feature selection, and the key importance of expert knowledge vs. automated selection. |
format | Online Article Text |
id | pubmed-8468242 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-84682422021-09-27 Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy Marschner, Sebastian N. Lombardo, Elia Minibek, Lena Holzgreve, Adrien Kaiser, Lena Albert, Nathalie L. Kurz, Christopher Riboldi, Marco Späth, Richard Baumeister, Philipp Niyazi, Maximilian Belka, Claus Corradini, Stefanie Landry, Guillaume Walter, Franziska Diagnostics (Basel) Article This study retrospectively analyzed the performance of artificial neural networks (ANN) to predict overall survival (OS) or locoregional failure (LRF) in HNSCC patients undergoing radiotherapy, based on 2-[(18)F]FDG PET/CT and clinical covariates. We compared predictions relying on three different sets of features, extracted from 230 patients. Specifically, (i) an automated feature selection method independent of expert rating was compared with (ii) clinical variables with proven influence on OS or LRF and (iii) clinical data plus expert-selected SUV metrics. The three sets were given as input to an artificial neural network for outcome prediction, evaluated by Harrell’s concordance index (HCI) and by testing stratification capability. For OS and LRF, the best performance was achieved with expert-based PET-features (0.71 HCI) and clinical variables (0.70 HCI), respectively. For OS stratification, all three feature sets were significant, whereas for LRF only expert-based PET-features successfully classified low vs. high-risk patients. Based on 2-[(18)F]FDG PET/CT features, stratification into risk groups using ANN for OS and LRF is possible. Differences in the results for different feature sets confirm the relevance of feature selection, and the key importance of expert knowledge vs. automated selection. MDPI 2021-08-31 /pmc/articles/PMC8468242/ /pubmed/34573924 http://dx.doi.org/10.3390/diagnostics11091581 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 Marschner, Sebastian N. Lombardo, Elia Minibek, Lena Holzgreve, Adrien Kaiser, Lena Albert, Nathalie L. Kurz, Christopher Riboldi, Marco Späth, Richard Baumeister, Philipp Niyazi, Maximilian Belka, Claus Corradini, Stefanie Landry, Guillaume Walter, Franziska Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title | Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title_full | Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title_fullStr | Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title_full_unstemmed | Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title_short | Risk Stratification Using (18)F-FDG PET/CT and Artificial Neural Networks in Head and Neck Cancer Patients Undergoing Radiotherapy |
title_sort | risk stratification using (18)f-fdg pet/ct and artificial neural networks in head and neck cancer patients undergoing radiotherapy |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8468242/ https://www.ncbi.nlm.nih.gov/pubmed/34573924 http://dx.doi.org/10.3390/diagnostics11091581 |
work_keys_str_mv | AT marschnersebastiann riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT lombardoelia riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT minibeklena riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT holzgreveadrien riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT kaiserlena riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT albertnathaliel riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT kurzchristopher riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT riboldimarco riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT spathrichard riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT baumeisterphilipp riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT niyazimaximilian riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT belkaclaus riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT corradinistefanie riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT landryguillaume riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy AT walterfranziska riskstratificationusing18ffdgpetctandartificialneuralnetworksinheadandneckcancerpatientsundergoingradiotherapy |