Haiqun Xing1,2, Zhixin Hao1,2, Wenjia Zhu1,2, Dehui Sun3, Jie Ding1,2, Hui Zhang4, Yu Liu1,2, Li Huo5,6. 1. Department of Nuclear Medicine, Peking Union Medical College Hospital, Dongcheng District, Chinese Academy of Medical Science, Peking Union Medical College, No.1 Shuaifuyuan, Beijing, 100730, China. 2. Beijing Key Laboratory of Molecular Targeted Diagnosis and Therapy in Nuclear Medicine, Beijing, 100730, China. 3. Sinounion Healthcare Inc., Building 3-B, Zhongguancun Dong Sheng International Pioneer Park, Beijing, 100192, China. 4. Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, 100084, China. 5. Department of Nuclear Medicine, Peking Union Medical College Hospital, Dongcheng District, Chinese Academy of Medical Science, Peking Union Medical College, No.1 Shuaifuyuan, Beijing, 100730, China. huoli@pumch.cn. 6. Beijing Key Laboratory of Molecular Targeted Diagnosis and Therapy in Nuclear Medicine, Beijing, 100730, China. huoli@pumch.cn.
Abstract
PURPOSE: To develop and validate a machine learning model based on radiomic features derived from 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) images to preoperatively predict the pathological grade in patients with pancreatic ductal adenocarcinoma (PDAC). METHODS: A total of 149 patients (83 men, 66 women, mean age 61 years old) with pathologically proven PDAC and a preoperative 18F-FDG PET/CT scan between May 2009 and January 2016 were included in this retrospective study. The cohort of patients was divided into two separate groups for the training (99 patients) and validation (50 patients) in chronological order. Radiomics features were extracted from PET/CT images using Pyradiomics implemented in Python, and the XGBoost algorithm was used to build a prediction model. Conventional PET parameters, including standardized uptake value, metabolic tumor volume, and total lesion glycolysis, were also measured. The quality of the proposed model was appraised by means of receiver operating characteristics (ROC) and areas under the ROC curve (AUC). RESULTS: The prediction model based on a twelve-feature-combined radiomics signature could stratify PDAC patients into grade 1 and grade 2/3 groups with AUC of 0.994 in the training set and 0.921 in the validation set. CONCLUSION: The model developed is capable of predicting pathological differentiation grade of PDAC based on preoperative 18F-FDG PET/CT radiomics features.
PURPOSE: To develop and validate a machine learning model based on radiomic features derived from 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) images to preoperatively predict the pathological grade in patients with pancreatic ductal adenocarcinoma (PDAC). METHODS: A total of 149 patients (83 men, 66 women, mean age 61 years old) with pathologically proven PDAC and a preoperative 18F-FDG PET/CT scan between May 2009 and January 2016 were included in this retrospective study. The cohort of patients was divided into two separate groups for the training (99 patients) and validation (50 patients) in chronological order. Radiomics features were extracted from PET/CT images using Pyradiomics implemented in Python, and the XGBoost algorithm was used to build a prediction model. Conventional PET parameters, including standardized uptake value, metabolic tumor volume, and total lesion glycolysis, were also measured. The quality of the proposed model was appraised by means of receiver operating characteristics (ROC) and areas under the ROC curve (AUC). RESULTS: The prediction model based on a twelve-feature-combined radiomics signature could stratify PDACpatients into grade 1 and grade 2/3 groups with AUC of 0.994 in the training set and 0.921 in the validation set. CONCLUSION: The model developed is capable of predicting pathological differentiation grade of PDAC based on preoperative 18F-FDG PET/CT radiomics features.
Authors: Alexander P Stark; Greg D Sacks; Matthew M Rochefort; Timothy R Donahue; Howard A Reber; James S Tomlinson; David W Dawson; Guido Eibl; O Joe Hines Journal: Surgery Date: 2016-02-02 Impact factor: 3.982
Authors: M Hidalgo; R Álvarez; J Gallego; C Guillén-Ponce; B Laquente; T Macarulla; A Muñoz; M Salgado; R Vera; J Adeva; I Alés; S Arévalo; J Blázquez; A Calsina; A Carmona; E de Madaria; R Díaz; L Díez; T Fernández; B G de Paredes; M E Gallardo; I González; O Hernando; P Jiménez; A López; C López; F López-Ríos; E Martín; J Martínez; A Martínez; J Montans; R Pazo; J C Plaza; I Peiró; J J Reina; A Sanjuanbenito; R Yaya; Alfredo Carrato Journal: Clin Transl Oncol Date: 2016-12-19 Impact factor: 3.405
Authors: Horacio J Asbun; Kevin Conlon; Laureano Fernandez-Cruz; Helmut Friess; Shailesh V Shrikhande; Mustapha Adham; Claudio Bassi; Maximilian Bockhorn; Markus Büchler; Richard M Charnley; Christos Dervenis; Abe Fingerhutt; Dirk J Gouma; Werner Hartwig; Clem Imrie; Jakob R Izbicki; Keith D Lillemoe; Miroslav Milicevic; Marco Montorsi; John P Neoptolemos; Aken A Sandberg; Michael Sarr; Charles Vollmer; Charles J Yeo; L William Traverso Journal: Surgery Date: 2014-01-02 Impact factor: 3.982
Authors: Maria Elena Laino; Angela Ammirabile; Ludovica Lofino; Lorenzo Mannelli; Francesco Fiz; Marco Francone; Arturo Chiti; Luca Saba; Matteo Agostino Orlandi; Victor Savevski Journal: Healthcare (Basel) Date: 2022-08-11