FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955875877> ?p ?o ?g. }

• W2955875877 endingPage "iv17" @default.
• W2955875877 startingPage "iv17" @default.
• W2955875877 abstract "Introduction: There are few tools that can be confidently used in current clinical practice to predict survival in esophageal cancer. In the literature, the discrimination ability of clinical tools based models was around 0.63 to 0.77, which was far from confident application in clinical use. We aimed to develop a deep convolutional neural network (CNN) based positron emission tomography (PET) image analysis to predict esophageal cancer outcome. Methods: A total of 798 PET scans of esophageal squamous cell carcinoma and 309 PET scans of stage I lung cancer, whose esophagus was presumed to be normal were collected. In the first stage, we pre-trained a 3D-CNN using all the PET scans, with the task to classify the scans into esophageal cancer or lung cancer. The PET scans of 548 esophageal cancer patients were included in the second stage with an aim to classify the patients who expired within or survived more than one year after diagnosis. The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance. Survival analysis was also performed to evaluate the impact of prediction results. Results: The deep CNN (pre-train/Adam optimizer/34-layer network model) attained an AUC of 0.738 in identifying patients who expired with one year after diagnosis. In the survival analysis, patients who were predicted to be expired but were alive at one year after diagnosis had a 5-year survival rate of 32.6%, which was significantly worse than those who were predicted to survive and really survived one year after diagnosis (50.5%, P < .001), suggesting that the prediction model could identify tumors with more aggressive behavior. In the multivariable analysis, the independent prognostic factors for overall survival included age, gender, tumor location at upper third, clinical N stage, clinical M stage, and the prediction result (HR 2.830; 95% CI, 2.252-3.555, P < .001). Conclusion: A 3D-CNN can be trained with PET imaging datasets to predict esophageal cancer outcome with acceptable accuracy. Although our current results cannot be readily applied to clinical decision-making, we demonstrated the potential of deep learning. With a larger dataset, CNN can be trained to achieve better prediction performance." @default.
• W2955875877 created "2019-07-12" @default.
• W2955875877 creator A5000223797 @default.
• W2955875877 creator A5055464489 @default.
• W2955875877 date "2019-07-01" @default.
• W2955875877 modified "2023-10-16" @default.
• W2955875877 title "Predicting esophageal cancer outcome with positron emission tomography using deep convolutional neural network" @default.
• W2955875877 doi "https://doi.org/10.1093/annonc/mdz155.061" @default.
• W2955875877 hasPublicationYear "2019" @default.
• W2955875877 type Work @default.
• W2955875877 sameAs 2955875877 @default.
• W2955875877 citedByCount "0" @default.
• W2955875877 crossrefType "journal-article" @default.
• W2955875877 hasAuthorship W2955875877A5000223797 @default.
• W2955875877 hasAuthorship W2955875877A5055464489 @default.
• W2955875877 hasBestOaLocation W29558758771 @default.
• W2955875877 hasConcept C108583219 @default.
• W2955875877 hasConcept C121608353 @default.
• W2955875877 hasConcept C126322002 @default.
• W2955875877 hasConcept C126838900 @default.
• W2955875877 hasConcept C143998085 @default.
• W2955875877 hasConcept C146357865 @default.
• W2955875877 hasConcept C151730666 @default.
• W2955875877 hasConcept C154945302 @default.
• W2955875877 hasConcept C2775842073 @default.
• W2955875877 hasConcept C2776256026 @default.
• W2955875877 hasConcept C2777297899 @default.
• W2955875877 hasConcept C2777819096 @default.
• W2955875877 hasConcept C2779742542 @default.
• W2955875877 hasConcept C2989005 @default.
• W2955875877 hasConcept C41008148 @default.
• W2955875877 hasConcept C58471807 @default.
• W2955875877 hasConcept C71924100 @default.
• W2955875877 hasConcept C81363708 @default.
• W2955875877 hasConcept C86803240 @default.
• W2955875877 hasConceptScore W2955875877C108583219 @default.
• W2955875877 hasConceptScore W2955875877C121608353 @default.
• W2955875877 hasConceptScore W2955875877C126322002 @default.
• W2955875877 hasConceptScore W2955875877C126838900 @default.
• W2955875877 hasConceptScore W2955875877C143998085 @default.
• W2955875877 hasConceptScore W2955875877C146357865 @default.
• W2955875877 hasConceptScore W2955875877C151730666 @default.
• W2955875877 hasConceptScore W2955875877C154945302 @default.
• W2955875877 hasConceptScore W2955875877C2775842073 @default.
• W2955875877 hasConceptScore W2955875877C2776256026 @default.
• W2955875877 hasConceptScore W2955875877C2777297899 @default.
• W2955875877 hasConceptScore W2955875877C2777819096 @default.
• W2955875877 hasConceptScore W2955875877C2779742542 @default.
• W2955875877 hasConceptScore W2955875877C2989005 @default.
• W2955875877 hasConceptScore W2955875877C41008148 @default.
• W2955875877 hasConceptScore W2955875877C58471807 @default.
• W2955875877 hasConceptScore W2955875877C71924100 @default.
• W2955875877 hasConceptScore W2955875877C81363708 @default.
• W2955875877 hasConceptScore W2955875877C86803240 @default.
• W2955875877 hasLocation W29558758771 @default.
• W2955875877 hasOpenAccess W2955875877 @default.
• W2955875877 hasPrimaryLocation W29558758771 @default.
• W2955875877 hasRelatedWork W171727350 @default.
• W2955875877 hasRelatedWork W1969961543 @default.
• W2955875877 hasRelatedWork W2046901075 @default.
• W2955875877 hasRelatedWork W2054375708 @default.
• W2955875877 hasRelatedWork W2084845148 @default.
• W2955875877 hasRelatedWork W2275543989 @default.
• W2955875877 hasRelatedWork W2411206711 @default.
• W2955875877 hasRelatedWork W2417932425 @default.
• W2955875877 hasRelatedWork W2897898837 @default.
• W2955875877 hasRelatedWork W2945733336 @default.
• W2955875877 hasVolume "30" @default.
• W2955875877 isParatext "false" @default.
• W2955875877 isRetracted "false" @default.
• W2955875877 magId "2955875877" @default.
• W2955875877 workType "article" @default.