Deep learning analysis using FDG-PET to predict treatment outcome in patients with oral cavity squamous cell carcinoma.
Adult
Aged
Aged, 80 and over
Algorithms
Deep Learning
Diagnosis, Computer-Assisted
/ methods
Disease-Free Survival
Female
Fluorodeoxyglucose F18
Glycolysis
Humans
Kaplan-Meier Estimate
Male
Middle Aged
Mouth Neoplasms
/ diagnostic imaging
Neoplasm Staging
Positron Emission Tomography Computed Tomography
Predictive Value of Tests
Reproducibility of Results
Retrospective Studies
Sensitivity and Specificity
Squamous Cell Carcinoma of Head and Neck
/ diagnostic imaging
Treatment Outcome
Tumor Burden
Deep learning
Disease-free survival
Fluorodeoxyglucose F18
Positron emission tomography
Squamous cell carcinoma of head and neck
Journal
European radiology
ISSN: 1432-1084
Titre abrégé: Eur Radiol
Pays: Germany
ID NLM: 9114774
Informations de publication
Date de publication:
Nov 2020
Nov 2020
Historique:
received:
20
04
2020
accepted:
26
05
2020
revised:
20
04
2020
pubmed:
12
6
2020
medline:
16
3
2021
entrez:
12
6
2020
Statut:
ppublish
Résumé
To assess the utility of deep learning analysis using One hundred thirteen patients with OCSCC who received pretreatment FDG-PET/CT were included. They were divided into training (83 patients) and test (30 patients) sets. The diagnosis of treatment control/failure and the DFS rate were obtained from patients' medical records. In deep learning analyses, three planes of axial, coronal, and sagittal FDG-PET images were assessed by ResNet-101 architecture. In the training set, image analysis was performed for the diagnostic model creation. The test data set was subsequently analyzed for confirmation of diagnostic accuracy. T-stage, clinical stage, and conventional FDG-PET parameters (the maximum and mean standardized uptake value (SUVmax and SUVmean), heterogeneity index, metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were also assessed with determining the optimal cutoff from training dataset and then validated their diagnostic ability from test dataset. In dividing into patients with treatment control and failure, the highest diagnostic accuracy of 0.8 was obtained using deep learning classification, with a sensitivity of 0.8, specificity of 0.8, positive predictive value of 0.89, and negative predictive value of 0.67. In the Kaplan-Meier analysis, the DFS rate was significantly different only with the analysis of deep learning-based classification (p < .01). Deep learning-based diagnosis with FDG-PET images may predict treatment outcome in patients with OCSCC. • Deep learning-based diagnosis of FDG-PET images showed the highest diagnostic accuracy to predict the treatment outcome in patients with oral cavity squamous cell carcinoma. • Deep learning-based diagnosis was shown to differentiate patients between good and poor disease-free survival more clearly than conventional T-stage, clinical stage, and conventional FDG-PET-based parameters.
Identifiants
pubmed: 32524219
doi: 10.1007/s00330-020-06982-8
pii: 10.1007/s00330-020-06982-8
doi:
Substances chimiques
Fluorodeoxyglucose F18
0Z5B2CJX4D
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
6322-6330Références
Montero PH, Patel SG (2015) Cancer of the oral cavity. Surg Oncol Clin N Am 24:491–508
doi: 10.1016/j.soc.2015.03.006
Cooper JS, Pajak TF, Forastiere AA et al (2004) Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 350:1937–1944
doi: 10.1056/NEJMoa032646
Cheng YJ, Tsai MH, Chiang CJ et al (2018) Adjuvant radiotherapy after curative surgery for oral cavity squamous cell carcinoma and treatment effect of timing and duration on outcome-A Taiwan Cancer Registry national database analysis. Cancer Med. https://doi.org/10.1002/cam4.1611
Pasha MA, Marcus C, Fakhry C, Kang H, Kiess AP, Subramaniam RM (2015) FDG PET/CT for management and assessing outcomes of squamous cell cancer of the oral cavity. AJR Am J Roentgenol 205:W150–W161
doi: 10.2214/AJR.14.13830
Suzuki H, Fukuyama R, Hasegawa Y et al (2009) Tumor thickness, depth of invasion, and Bcl-2 expression are correlated with FDG-uptake in oral squamous cell carcinomas. Oral Oncol 45:891–897
doi: 10.1016/j.oraloncology.2009.03.009
Liao CT, Chang JT, Wang HM et al (2009) Pretreatment primary tumor SUVmax measured by FDG-PET and pathologic tumor depth predict for poor outcomes in patients with oral cavity squamous cell carcinoma and pathologically positive lymph nodes. Int J Radiat Oncol Biol Phys 73:764–771
doi: 10.1016/j.ijrobp.2008.05.004
Liao CT, Hsieh CH, Fan WL et al (2020) A combined analysis of maximum standardized uptake value on FDG-PET, genetic markers, and clinicopathological risk factors in the prognostic stratification of patients with resected oral cavity squamous cell carcinoma. Eur J Nucl Med Mol Imaging 47:84–93
doi: 10.1007/s00259-019-04453-x
Hasegawa O, Satomi T, Kono M, Watanabe M, Ikehata N, Chikazu D (2019) Correlation between the malignancy and prognosis of oral squamous cell carcinoma in the maximum standardized uptake value. Odontology 107:237–243
doi: 10.1007/s10266-018-0379-9
Zhang H, Seikaly H, Abele JT, Jeffery DT, Harris JR, O'Connell DA (2014) Metabolic tumour volume as a prognostic factor for oral cavity squamous cell carcinoma treated with primary surgery. J Otolaryngol Head Neck Surg 43:33
pubmed: 25312990
pmcid: 4198685
Zhang H, Seikaly H, Nguyen NT et al (2016) Validation of metabolic tumor volume as a prognostic factor for oral cavity squamous cell carcinoma treated with primary surgery. Oral Oncol 57:6–14
doi: 10.1016/j.oraloncology.2016.03.013
Abd El-Hafez YG, Moustafa HM, Khalil HF, Liao CT, Yen TC (2013) Total lesion glycolysis: a possible new prognostic parameter in oral cavity squamous cell carcinoma. Oral Oncol 49:261–268
doi: 10.1016/j.oraloncology.2012.09.005
Choi WR, Oh JS, Roh JL et al (2019) Metabolic tumor volume and total lesion glycolysis predict tumor progression and survival after salvage surgery for recurrent oral cavity squamous cell carcinoma. Head Neck 41:1846–1853
doi: 10.1002/hed.25622
Lee SJ, Choi JY, Lee HJ et al (2012) Prognostic value of volume-based (18)F-fluorodeoxyglucose PET/CT parameters in patients with clinically node-negative oral tongue squamous cell carcinoma. Korean J Radiol 13:752–759
doi: 10.3348/kjr.2012.13.6.752
Kimura M, Kato I, Ishibashi K et al (2019) The prognostic significance of intratumoral heterogeneity of 18F-FDG uptake in patients with oral cavity squamous cell carcinoma. Eur J Radiol 114:99–104
doi: 10.1016/j.ejrad.2019.03.004
Kwon SH, Yoon JK, An YS et al (2014) Prognostic significance of the intratumoral heterogeneity of (18) F-FDG uptake in oral cavity cancer. J Surg Oncol 110:702–706
doi: 10.1002/jso.23703
Cheng NM, Fang YH, Lee LY et al (2015) Zone-size nonuniformity of 18F-FDG PET regional textural features predicts survival in patients with oropharyngeal cancer. Eur J Nucl Med Mol Imaging 42:419–428
doi: 10.1007/s00259-014-2933-1
Cheng NM, Fang YD, Tsan DL et al (2018) Heterogeneity and irregularity of pretreatment (18)F-fluorodeoxyglucose positron emission tomography improved prognostic stratification of p16-negative high-risk squamous cell carcinoma of the oropharynx. Oral Oncol 78:156–162
doi: 10.1016/j.oraloncology.2018.01.030
Fujima N, Hirata K, Shiga T et al (2018) Integrating quantitative morphological and intratumoural textural characteristics in FDG-PET for the prediction of prognosis in pharynx squamous cell carcinoma patients. Clin Radiol 73:1059 e1051–1059 e1058
doi: 10.1016/j.crad.2018.08.011
Soffer S, Ben-Cohen A, Shimon O, Amitai MM, Greenspan H, Klang E (2019) Convolutional neural networks for radiologic images: a radiologist’s guide. Radiology 290:590–606
doi: 10.1148/radiol.2018180547
Diamant A, Chatterjee A, Vallieres M, Shenouda G, Seuntjens J (2019) Deep learning in head & neck cancer outcome prediction. Sci Rep 9:2764
doi: 10.1038/s41598-019-39206-1
Lim R, Eaton A, Lee NY et al (2012) 18F-FDG PET/CT metabolic tumor volume and total lesion glycolysis predict outcome in oropharyngeal squamous cell carcinoma. J Nucl Med 53:1506–1513
doi: 10.2967/jnumed.111.101402
Yasaka K, Akai H, Kunimatsu A, Kiryu S, Abe O (2018) Deep learning with convolutional neural network in radiology. Jpn J Radiol 36:257–272
doi: 10.1007/s11604-018-0726-3
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. Proc IEEE Conf Comput Vis Pattern Recognit 770–778
Apostolova I, Steffen IG, Wedel F et al (2014) Asphericity of pretherapeutic tumour FDG uptake provides independent prognostic value in head-and-neck cancer. Eur Radiol 24:2077–2087
doi: 10.1007/s00330-014-3269-8
Kim M, Higuchi T, Nakajima T et al (2019) (18)F-FDG and (18)F-FAMT PET-derived metabolic parameters predict outcome of oral squamous cell carcinoma. Oral Radiol 35:308–314
doi: 10.1007/s11282-019-00377-2
Giraud P, Gasnier A, El Ayachy R et al (2019) Radiomics and machine learning for radiotherapy in head and neck cancers. Front Oncol 9:174
doi: 10.3389/fonc.2019.00174
Buch K, Li B, Qureshi MM, Kuno H, Anderson SW, Sakai O (2017) Quantitative assessment of variation in CT parameters on texture features: pilot study using a nonanatomic phantom. AJNR Am J Neuroradiol 38:981–985
doi: 10.3174/ajnr.A5139
Buch K, Kuno H, Qureshi MM, Li B, Sakai O (2018) Quantitative variations in texture analysis features dependent on MRI scanning parameters: A phantom model. J Appl Clin Med Phys 19:253–264
doi: 10.1002/acm2.12482