Wei-Chih Shen1, Shang-Wen Chen2,3,4, Kuo-Chen Wu1, Te-Chun Hsieh5,6, Ji-An Liang2,7, Yao-Ching Hung3,8, Lian-Shung Yeh3,8, Wei-Chun Chang3,8, Wu-Chou Lin3,8, Kuo-Yang Yen5,6, Chia-Hung Kao9,10,11. 1. Department of Computer Science and Information Engineering, Asia University, Taichung, Taiwan. 2. Department of Radiation Oncology, China Medical University Hospital, Taichung, Taiwan. 3. School of Medicine, College of Medicine, China Medical University, Taichung, Taiwan. 4. Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan. 5. Department of Nuclear Medicine and PET Center, China Medical University Hospital, Taichung, Taiwan. 6. Department of Biomedical Imaging and Radiological Science, China Medical University, Taichung, Taiwan. 7. Graduate Institute of Biomedical Sciences, School of Medicine, College of Medicine, China Medical University, No. 2, Yuh-Der Road, Taichung, 404, Taiwan. 8. Department of Obstetrics and Gynecology, China Medical University Hospital, Taichung, Taiwan. 9. Department of Nuclear Medicine and PET Center, China Medical University Hospital, Taichung, Taiwan. d10040@mail.cmuh.org.tw. 10. Graduate Institute of Biomedical Sciences, School of Medicine, College of Medicine, China Medical University, No. 2, Yuh-Der Road, Taichung, 404, Taiwan. d10040@mail.cmuh.org.tw. 11. Department of Bioinformatics and Medical Engineering, Asia University, Taichung, Taiwan. d10040@mail.cmuh.org.tw.
Abstract
BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSION: 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. KEY POINTS: • 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.
BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSION: 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 cancerpatients. KEY POINTS: • 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 cancerpatients. • 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.
Authors: Dominique Delbeke; R Edward Coleman; Milton J Guiberteau; Manuel L Brown; Henry D Royal; Barry A Siegel; David W Townsend; Lincoln L Berland; J Anthony Parker; Karl Hubner; Michael G Stabin; George Zubal; Marc Kachelriess; Valerie Cronin; Scott Holbrook Journal: J Nucl Med Date: 2006-05 Impact factor: 10.057
Authors: Michiel Kallenberg; Kersten Petersen; Mads Nielsen; Andrew Y Ng; Christian Igel; Celine M Vachon; Katharina Holland; Rikke Rass Winkel; Nico Karssemeijer; Martin Lillholm Journal: IEEE Trans Med Imaging Date: 2016-02-18 Impact factor: 10.048
Authors: Gabriel Chartrand; Phillip M Cheng; Eugene Vorontsov; Michal Drozdzal; Simon Turcotte; Christopher J Pal; Samuel Kadoury; An Tang Journal: Radiographics Date: 2017 Nov-Dec Impact factor: 5.333
Authors: Nalee Kim; Won Park; Won Kyung Cho; Duk-Soo Bae; Byoung-Gie Kim; Jeong-Won Lee; Tae-Joong Kim; Chel Hun Choi; Yoo-Young Lee; Young Seok Cho Journal: Cancer Res Treat Date: 2020-12-09 Impact factor: 4.679
Authors: Marta Ferreira; Pierre Lovinfosse; Johanne Hermesse; Marjolein Decuypere; Caroline Rousseau; François Lucia; Ulrike Schick; Caroline Reinhold; Philippe Robin; Mathieu Hatt; Dimitris Visvikis; Claire Bernard; Ralph T H Leijenaar; Frédéric Kridelka; Philippe Lambin; Patrick E Meyer; Roland Hustinx Journal: Eur J Nucl Med Mol Imaging Date: 2021-03-26 Impact factor: 9.236