Prediction of the local treatment outcome in patients with oropharyngeal squamous cell carcinoma using deep learning analysis of pretreatment FDG-PET images.
Adult
Aged
Aged, 80 and over
Biomarkers, Tumor
Clinical Decision-Making
Combined Modality Therapy
Deep Learning
Disease Management
Female
Fluorodeoxyglucose F18
Humans
Image Processing, Computer-Assisted
Kaplan-Meier Estimate
Male
Middle Aged
Neoplasm Staging
Oropharyngeal Neoplasms
/ diagnosis
Positron Emission Tomography Computed Tomography
Positron-Emission Tomography
/ methods
Prognosis
ROC Curve
Squamous Cell Carcinoma of Head and Neck
/ diagnosis
Treatment Outcome
Workflow
Deep learning
FDG-PET
Oropharyngeal squamous cell carcinoma
Treatment outcome
Journal
BMC cancer
ISSN: 1471-2407
Titre abrégé: BMC Cancer
Pays: England
ID NLM: 100967800
Informations de publication
Date de publication:
06 Aug 2021
06 Aug 2021
Historique:
received:
12
01
2021
accepted:
09
07
2021
entrez:
7
8
2021
pubmed:
8
8
2021
medline:
21
10
2021
Statut:
epublish
Résumé
This study aimed to assess the utility of deep learning analysis using pretreatment FDG-PET images to predict local treatment outcome in oropharyngeal squamous cell carcinoma (OPSCC) patients. One hundred fifty-four OPSCC patients who received pretreatment FDG-PET were included and divided into training (n = 102) and test (n = 52) sets. The diagnosis of local failure and local progression-free survival (PFS) rates were obtained from patient medical records. In deep learning analyses, axial and coronal images were assessed by three different architectures (AlexNet, GoogLeNET, and ResNet). In the training set, FDG-PET images were analyzed after the data augmentation process for the diagnostic model creation. A multivariate clinical model was also created using a binomial logistic regression model from a patient's clinical characteristics. The test data set was subsequently analyzed for confirmation of diagnostic accuracy. Assessment of local PFS rates was also performed. Training sessions were successfully performed with an accuracy of 74-89%. ROC curve analyses revealed an AUC of 0.61-0.85 by the deep learning model in the test set, whereas it was 0.62 by T-stage, 0.59 by clinical stage, and 0.74 by a multivariate clinical model. The highest AUC (0.85) was obtained with deep learning analysis of ResNet architecture. Cox proportional hazards regression analysis revealed deep learning-based classification by a multivariate clinical model (P < .05), and ResNet (P < .001) was a significant predictor of the treatment outcome. In the Kaplan-Meier analysis, the deep learning-based classification divided the patient's local PFS rate better than the T-stage, clinical stage, and a multivariate clinical model. Deep learning-based diagnostic model with FDG-PET images indicated its possibility to predict local treatment outcomes in OPSCCs.
Sections du résumé
BACKGROUND
BACKGROUND
This study aimed to assess the utility of deep learning analysis using pretreatment FDG-PET images to predict local treatment outcome in oropharyngeal squamous cell carcinoma (OPSCC) patients.
METHODS
METHODS
One hundred fifty-four OPSCC patients who received pretreatment FDG-PET were included and divided into training (n = 102) and test (n = 52) sets. The diagnosis of local failure and local progression-free survival (PFS) rates were obtained from patient medical records. In deep learning analyses, axial and coronal images were assessed by three different architectures (AlexNet, GoogLeNET, and ResNet). In the training set, FDG-PET images were analyzed after the data augmentation process for the diagnostic model creation. A multivariate clinical model was also created using a binomial logistic regression model from a patient's clinical characteristics. The test data set was subsequently analyzed for confirmation of diagnostic accuracy. Assessment of local PFS rates was also performed.
RESULTS
RESULTS
Training sessions were successfully performed with an accuracy of 74-89%. ROC curve analyses revealed an AUC of 0.61-0.85 by the deep learning model in the test set, whereas it was 0.62 by T-stage, 0.59 by clinical stage, and 0.74 by a multivariate clinical model. The highest AUC (0.85) was obtained with deep learning analysis of ResNet architecture. Cox proportional hazards regression analysis revealed deep learning-based classification by a multivariate clinical model (P < .05), and ResNet (P < .001) was a significant predictor of the treatment outcome. In the Kaplan-Meier analysis, the deep learning-based classification divided the patient's local PFS rate better than the T-stage, clinical stage, and a multivariate clinical model.
CONCLUSIONS
CONCLUSIONS
Deep learning-based diagnostic model with FDG-PET images indicated its possibility to predict local treatment outcomes in OPSCCs.
Identifiants
pubmed: 34362317
doi: 10.1186/s12885-021-08599-6
pii: 10.1186/s12885-021-08599-6
pmc: PMC8344209
doi:
Substances chimiques
Biomarkers, Tumor
0
Fluorodeoxyglucose F18
0Z5B2CJX4D
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
900Informations de copyright
© 2021. The Author(s).
Références
Oral Oncol. 2018 Mar;78:156-162
pubmed: 29496044
AJNR Am J Neuroradiol. 2017 Dec;38(12):2334-2340
pubmed: 29025727
Radiology. 2017 Dec;285(3):713-718
pubmed: 29155639
Eur J Radiol. 2020 May;126:108936
pubmed: 32171912
Eur J Nucl Med Mol Imaging. 2015 Mar;42(3):419-28
pubmed: 25339524
Int J Clin Oncol. 2017 Aug;22(4):682-689
pubmed: 28271391
Eur Radiol. 2014 Sep;24(9):2077-87
pubmed: 24965509
Cancers (Basel). 2019 Jun 10;11(6):
pubmed: 31185611
Sci Rep. 2019 Feb 26;9(1):2764
pubmed: 30809047
Eur Radiol. 2020 Nov;30(11):6311-6321
pubmed: 32500196
Sci Rep. 2018 Jan 24;8(1):1524
pubmed: 29367653
JCO Clin Cancer Inform. 2018 Dec;2:1-19
pubmed: 30652615
Medicine (Baltimore). 2019 May;98(18):e15446
pubmed: 31045814
Eur Radiol. 2020 Nov;30(11):6322-6330
pubmed: 32524219
J Nucl Med. 2012 Oct;53(10):1506-13
pubmed: 22895812
Int J Radiat Oncol Biol Phys. 2012 Feb 1;82(2):548-53
pubmed: 21277108
Jpn J Radiol. 2018 Apr;36(4):257-272
pubmed: 29498017
Head Neck. 2013 Jan;35(1):15-22
pubmed: 22307893
N Engl J Med. 2010 Jul 1;363(1):24-35
pubmed: 20530316
Clin Radiol. 2018 Dec;73(12):1059.e1-1059.e8
pubmed: 30245069
Phys Med Biol. 2017 Jul 7;62(13):5327-5343
pubmed: 28604368
Sci Rep. 2020 Jan 16;10(1):526
pubmed: 31949181
Cancer. 2011 Apr 15;117(8):1679-86
pubmed: 21472715
Front Oncol. 2017 Jun 30;7:129
pubmed: 28713770
JAMA Otolaryngol Head Neck Surg. 2016 Jan;142(1):79-85
pubmed: 26583955