Matches in SemOpenAlex for { <https://semopenalex.org/work/W4296193742> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4296193742 endingPage "135" @default.
- W4296193742 startingPage "126" @default.
- W4296193742 abstract "H &E images can be utilized to predict genetic mutations as biomarkers to potentially substitute many molecular biomarker assays in order to aid patients. Having a single model built by conducting prediction tasks simultaneously can save computation resources and provide a more generalizable model for future usage. A basic technique for generating such a comprehensive and efficient model is to employ a multi-task learning approach. However, overfitting the model to the trivial answers can occur in training for multiple tasks with extremely imbalanced class labels where resampling and rebalancing for all minor classes simultaneously are prohibited. Herein we propose a sequential multi-task learning approach to train a single model capable of predicting multiple genetic mutations while avoiding overfitting to trivial answers for imbalanced classes. We compared our strategy to the baseline multi-task training, as well as two more advanced approaches: (1) using weighted loss and (2) using self-supervised pre-training. We also used a trimming method to deal with noisy labels. To assess our methods, we trained models to predict 10 genetic mutations on the H &E images of the TCGA-LUAD dataset. AUROC and F1 score are reported, while we demonstrate that F1 score may be a more suitable metric for multi-task learning with imbalanced labels. It is shown that our proposed trimming strategy combined with sequential learning could improve the predictions on all of the mutations compared with other multi-task learning approaches. Also, we investigated the application of continual learning." @default.
- W4296193742 created "2022-09-18" @default.
- W4296193742 creator A5036328196 @default.
- W4296193742 creator A5040630873 @default.
- W4296193742 creator A5041070280 @default.
- W4296193742 creator A5065172378 @default.
- W4296193742 creator A5084345521 @default.
- W4296193742 creator A5085529882 @default.
- W4296193742 creator A5085822711 @default.
- W4296193742 date "2022-01-01" @default.
- W4296193742 modified "2023-10-15" @default.
- W4296193742 title "Sequential Multi-task Learning for Histopathology-Based Prediction of Genetic Mutations with Extremely Imbalanced Labels" @default.
- W4296193742 cites W2473930607 @default.
- W4296193742 cites W2754791538 @default.
- W4296193742 cites W2760946358 @default.
- W4296193742 cites W2788388592 @default.
- W4296193742 cites W3036901136 @default.
- W4296193742 cites W3043835773 @default.
- W4296193742 cites W3044996171 @default.
- W4296193742 cites W3098750840 @default.
- W4296193742 cites W3126827997 @default.
- W4296193742 cites W3171888599 @default.
- W4296193742 cites W3174917907 @default.
- W4296193742 cites W3212889265 @default.
- W4296193742 cites W4309477120 @default.
- W4296193742 cites W4312648273 @default.
- W4296193742 doi "https://doi.org/10.1007/978-3-031-16961-8_13" @default.
- W4296193742 hasPublicationYear "2022" @default.
- W4296193742 type Work @default.
- W4296193742 citedByCount "0" @default.
- W4296193742 crossrefType "book-chapter" @default.
- W4296193742 hasAuthorship W4296193742A5036328196 @default.
- W4296193742 hasAuthorship W4296193742A5040630873 @default.
- W4296193742 hasAuthorship W4296193742A5041070280 @default.
- W4296193742 hasAuthorship W4296193742A5065172378 @default.
- W4296193742 hasAuthorship W4296193742A5084345521 @default.
- W4296193742 hasAuthorship W4296193742A5085529882 @default.
- W4296193742 hasAuthorship W4296193742A5085822711 @default.
- W4296193742 hasConcept C111919701 @default.
- W4296193742 hasConcept C119857082 @default.
- W4296193742 hasConcept C150921843 @default.
- W4296193742 hasConcept C154945302 @default.
- W4296193742 hasConcept C162324750 @default.
- W4296193742 hasConcept C176217482 @default.
- W4296193742 hasConcept C187736073 @default.
- W4296193742 hasConcept C21547014 @default.
- W4296193742 hasConcept C22019652 @default.
- W4296193742 hasConcept C2780451532 @default.
- W4296193742 hasConcept C41008148 @default.
- W4296193742 hasConcept C50644808 @default.
- W4296193742 hasConcept C56951928 @default.
- W4296193742 hasConcept C8880873 @default.
- W4296193742 hasConceptScore W4296193742C111919701 @default.
- W4296193742 hasConceptScore W4296193742C119857082 @default.
- W4296193742 hasConceptScore W4296193742C150921843 @default.
- W4296193742 hasConceptScore W4296193742C154945302 @default.
- W4296193742 hasConceptScore W4296193742C162324750 @default.
- W4296193742 hasConceptScore W4296193742C176217482 @default.
- W4296193742 hasConceptScore W4296193742C187736073 @default.
- W4296193742 hasConceptScore W4296193742C21547014 @default.
- W4296193742 hasConceptScore W4296193742C22019652 @default.
- W4296193742 hasConceptScore W4296193742C2780451532 @default.
- W4296193742 hasConceptScore W4296193742C41008148 @default.
- W4296193742 hasConceptScore W4296193742C50644808 @default.
- W4296193742 hasConceptScore W4296193742C56951928 @default.
- W4296193742 hasConceptScore W4296193742C8880873 @default.
- W4296193742 hasLocation W42961937421 @default.
- W4296193742 hasOpenAccess W4296193742 @default.
- W4296193742 hasPrimaryLocation W42961937421 @default.
- W4296193742 hasRelatedWork W1996541855 @default.
- W4296193742 hasRelatedWork W2985459377 @default.
- W4296193742 hasRelatedWork W2989932438 @default.
- W4296193742 hasRelatedWork W3011996705 @default.
- W4296193742 hasRelatedWork W3036030544 @default.
- W4296193742 hasRelatedWork W3099765033 @default.
- W4296193742 hasRelatedWork W3175189414 @default.
- W4296193742 hasRelatedWork W4210794429 @default.
- W4296193742 hasRelatedWork W4213073923 @default.
- W4296193742 hasRelatedWork W4296193742 @default.
- W4296193742 isParatext "false" @default.
- W4296193742 isRetracted "false" @default.
- W4296193742 workType "book-chapter" @default.