Prediction of local relapse and distant metastasis in patients with definitive chemoradiotherapy-treated cervical cancer by deep learning from [18F]-fluorodeoxyglucose positron emission tomography/computed tomography
- 196 Downloads
We designed a deep learning model for assessing 18F-FDG PET/CT for early prediction of local and distant failures for patients with locally advanced cervical cancer.
All 142 patients with cervical cancer underwent 18F-FDG PET/CT for pretreatment staging and received allocated treatment. To augment the amount of image data, each tumor was represented as 11 slice sets each of which contains 3 2D orthogonal slices to acquire a total of 1562 slice sets. In each round of k-fold cross-validation, a well-trained proposed model and a slice-based optimal threshold were derived from a training set and used to classify each slice set in the test set into the categories of with or without local or distant failure. The classification results of each tumor were aggregated to summarize a tumor-based prediction result.
In total, 21 and 26 patients experienced local and distant failures, respectively. Regarding local recurrence, the tumor-based prediction result summarized from all test sets demonstrated that the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 71%, 93%, 63%, 95%, and 89%, respectively. The corresponding values for distant metastasis were 77%, 90%, 63%, 95%, and 87%, respectively.
This is the first study to use deep learning model for assessing 18F-FDG PET/CT images which is capable of predicting treatment outcomes in cervical cancer patients.
• This is the first study to use deep learning model for assessing 18 F-FDG PET/CT images which is capable of predicting treatment outcomes in cervical cancer patients.
• All 142 patients with cervical cancer underwent 18 F-FDG PET/CT for pretreatment staging and received allocated treatment. To augment the amount of image data, each tumor was represented as 11 slice sets each of which contains 3 2D orthogonal slices to acquire a total of 1562 slice sets.
• For local recurrence, all test sets demonstrated that the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 71%, 93%, 63%, 95%, and 89%, respectively. The corresponding values for distant metastasis were 77%, 90%, 63%, 95%, and 87%, respectively.
KeywordsAI (artificial intelligence) Neural network models Fluorodeoxyglucose F18 PET-CT scan Cervical cancer
- 18F-FDG PET/CT
18F-fluorodeoxyglucose positron emission tomography-computed tomography
Clinical target volume
High gray-level run emphasis
Metabolic tumor volume
Pelvic lymph nodes
Standardized uptake value
Volume of interest
This work was supported by grants from the Ministry of Health and Welfare, Taiwan (MOHW107-TDU-B-212-123004); China Medical University Hospital (DMR-107-192, CRS-106-036, CRS106-039, CRS106-040, CRS106-041); Asia University (DMR-106-150); Academia Sinica Stroke Biosignature Project (BM10701010021); MOST Clinical Trial Consortium for Stroke (MOST 107-2321-B-039-004-); Tseng-Lien Lin Foundation, Taichung, Taiwan; and Katsuzo and Kiyo Aoshima Memorial Funds, Japan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding was received for this study.
Compliance with ethical standards
The scientific guarantor of this publication is Chia-Hung Kao, MD, Graduate Institute of Biomedical Sciences and School of Medicine, College of Medicine, China Medical University, No. 2, Yuh-Der Road, Taichung 404, Taiwan. E-mail: email@example.com; firstname.lastname@example.org
Conflict of interest
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Statistics and biometry
One of the authors (Prof. Shang-Wen Chen) has significant statistical expertise.
This is a retrospective study for images’ analyses. The IRB also specifically waived the consent requirement.
This study was approved by the local institutional review board (certificate numbers CMUH102-REC2-74 and DMR99-IRB-010-1).
Study subjects or cohorts overlap
This work was partially presented at NVIDIA GTC Taiwan 2018 Poster Contest.
• Diagnostic or prognostic study
• Performed at one institution
- 1.Rose PG, Java J, Whitney CW et al (2015) Nomograms predicting progression-free survival, overall survival, and pelvic recurrence in locally advanced cervical cancer developed from an analysis of identifiable prognostic pactors in patients from NRG Oncology/Gynecologic Oncology Group randomized trials of chemoradiotherapy. J Clin Oncol 33:2136–2142CrossRefGoogle Scholar
- 10.Ho KC, Fang YH, Chung HW et al (2016) A preliminary investigation into textural features of intratumoral metabolic heterogeneity in FDG PET for overall survival prognosis in patients with bulky cervical cancer treated with definitive concurrent chemoradiotherapy. Am J Nucl Med Mol Imaging 6:166–175PubMedPubMedCentralGoogle Scholar
- 16.Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105Google Scholar
- 27.Lin M, Chen Q, Yan S (2014) Network In Network. arXiv:1312.4400 [cs.NE] Available via https://arxiv.org/abs/1312.4400
- 28.Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167 [cs.LG] Avaliable via https://arxiv.org/abs/1502.03167
- 29.Chen SW, Liang JA, Hung YC et al (2013) Does initial 45Gy of pelvic intensity-modulated radiotherapy reduce late complications in patients with locally advanced cervical cancer? A cohort control study using definitive chemoradiotherapy with high-dose rate brachytherapy. Radiol Oncol 47:176–184CrossRefGoogle Scholar
- 30.Pötter R, Haie-Meder C, Van Limbergen E et al (2006) Recommendations from gynaecological (GYN) GEC ESTRO working group (II): concepts and terms in 3D image-based treatment planning in cervix cancer brachytherapy-3D dose volume parameters and aspects of 3D image-based anatomy, radiation physics, radiobiology. Radiother Oncol 78:67–77CrossRefGoogle Scholar