Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313457532> ?p ?o ?g. }
- W4313457532 abstract "Abstract Purpose To evaluate the impact of harmonization and oversampling methods on multi-center imbalanced datasets, with specific application to PET-based radiomics modeling for histologic subtype prediction in non-small cell lung cancer (NSCLC). Methods The study included 245 patients with adenocarcinoma (ADC) and 78 patients with squamous cell carcinoma (SCC) from 4 centers. Utilizing 1502 radiomics features per patient, we trained, validated, and externally tested 4 machine-learning classifiers, to investigate the effect of no harmonization (NoH) or 4 harmonization methods, paired with no oversampling (NoO) or 5 oversampling methods on subtype prediction. Model performance was evaluated using the average area under the ROC curve (AUROC) and G-mean via 5 times 5-fold cross-validations. Statistical comparisons of the combined models against baseline (NoH+NoO) were performed for each fold of cross-validation using the DeLong test. Results The number of cross-combinations with both AUROC and G-mean outperforming baseline in internal validation and external testing was 15, 4, 2, and 7 (out of 29) for random forest (RF), linear discriminant analysis (LDA), logistic regression (LR), and support vector machine (SVM), respectively. ComBat harmonization combined with oversampling (SMOTE) via RF yielded better performance than baseline (AUROC and G-mean of internal validation: 0.725 vs. 0.608 and 0.625 vs. 0.398; external testing: 0.637 vs. 0.567 and 0.363 vs. 0.234), though statistical significances were not observed. Conclusion Applying harmonization and oversampling methods in multi-center imbalanced datasets can improve NSCLC-subtype prediction, but varies widely across classifiers. We have created open-source comparisons of harmonization and oversampling on different classifiers for comprehensive evaluations in different studies." @default.
- W4313457532 created "2023-01-06" @default.
- W4313457532 creator A5007891293 @default.
- W4313457532 creator A5021438906 @default.
- W4313457532 creator A5026616191 @default.
- W4313457532 creator A5031854562 @default.
- W4313457532 creator A5034625862 @default.
- W4313457532 creator A5044928666 @default.
- W4313457532 creator A5054554103 @default.
- W4313457532 creator A5058744346 @default.
- W4313457532 creator A5059101247 @default.
- W4313457532 creator A5063110838 @default.
- W4313457532 date "2023-01-03" @default.
- W4313457532 modified "2023-09-29" @default.
- W4313457532 title "Impact of harmonization and oversampling methods on radiomics analysis of multi-center imbalanced datasets: Application to PET-based prediction of lung cancer subtypes" @default.
- W4313457532 cites W1987971958 @default.
- W4313457532 cites W2083927153 @default.
- W4313457532 cites W2118978333 @default.
- W4313457532 cites W2130681687 @default.
- W4313457532 cites W2139939379 @default.
- W4313457532 cites W2148143831 @default.
- W4313457532 cites W2154053567 @default.
- W4313457532 cites W2166401924 @default.
- W4313457532 cites W2167277498 @default.
- W4313457532 cites W2327037637 @default.
- W4313457532 cites W2328176404 @default.
- W4313457532 cites W2473959085 @default.
- W4313457532 cites W2520586201 @default.
- W4313457532 cites W2767128594 @default.
- W4313457532 cites W2781993955 @default.
- W4313457532 cites W2894753707 @default.
- W4313457532 cites W2896246551 @default.
- W4313457532 cites W2954996726 @default.
- W4313457532 cites W3007204104 @default.
- W4313457532 cites W3014904955 @default.
- W4313457532 cites W3017712284 @default.
- W4313457532 cites W3044994436 @default.
- W4313457532 cites W3048173156 @default.
- W4313457532 cites W3084293538 @default.
- W4313457532 cites W3092481189 @default.
- W4313457532 cites W3118615436 @default.
- W4313457532 cites W3128646645 @default.
- W4313457532 cites W3130746258 @default.
- W4313457532 cites W3135096391 @default.
- W4313457532 cites W3137777844 @default.
- W4313457532 cites W3167485041 @default.
- W4313457532 cites W3177078060 @default.
- W4313457532 cites W3193956823 @default.
- W4313457532 cites W3199011255 @default.
- W4313457532 cites W3199368218 @default.
- W4313457532 cites W3200827899 @default.
- W4313457532 cites W3208483636 @default.
- W4313457532 cites W3212319976 @default.
- W4313457532 cites W4200427670 @default.
- W4313457532 cites W4205334579 @default.
- W4313457532 cites W4210249522 @default.
- W4313457532 cites W4220981043 @default.
- W4313457532 cites W4282913119 @default.
- W4313457532 cites W4282979895 @default.
- W4313457532 cites W4283313052 @default.
- W4313457532 cites W4288060250 @default.
- W4313457532 doi "https://doi.org/10.21203/rs.3.rs-2393890/v1" @default.
- W4313457532 hasPublicationYear "2023" @default.
- W4313457532 type Work @default.
- W4313457532 citedByCount "0" @default.
- W4313457532 crossrefType "posted-content" @default.
- W4313457532 hasAuthorship W4313457532A5007891293 @default.
- W4313457532 hasAuthorship W4313457532A5021438906 @default.
- W4313457532 hasAuthorship W4313457532A5026616191 @default.
- W4313457532 hasAuthorship W4313457532A5031854562 @default.
- W4313457532 hasAuthorship W4313457532A5034625862 @default.
- W4313457532 hasAuthorship W4313457532A5044928666 @default.
- W4313457532 hasAuthorship W4313457532A5054554103 @default.
- W4313457532 hasAuthorship W4313457532A5058744346 @default.
- W4313457532 hasAuthorship W4313457532A5059101247 @default.
- W4313457532 hasAuthorship W4313457532A5063110838 @default.
- W4313457532 hasBestOaLocation W43134575321 @default.
- W4313457532 hasConcept C119857082 @default.
- W4313457532 hasConcept C121332964 @default.
- W4313457532 hasConcept C12267149 @default.
- W4313457532 hasConcept C151956035 @default.
- W4313457532 hasConcept C154945302 @default.
- W4313457532 hasConcept C169258074 @default.
- W4313457532 hasConcept C197323446 @default.
- W4313457532 hasConcept C24890656 @default.
- W4313457532 hasConcept C2776257435 @default.
- W4313457532 hasConcept C2778217198 @default.
- W4313457532 hasConcept C2778559731 @default.
- W4313457532 hasConcept C2779962950 @default.
- W4313457532 hasConcept C31258907 @default.
- W4313457532 hasConcept C41008148 @default.
- W4313457532 hasConcept C62520636 @default.
- W4313457532 hasConcept C69738355 @default.
- W4313457532 hasConcept C71924100 @default.
- W4313457532 hasConceptScore W4313457532C119857082 @default.
- W4313457532 hasConceptScore W4313457532C121332964 @default.
- W4313457532 hasConceptScore W4313457532C12267149 @default.
- W4313457532 hasConceptScore W4313457532C151956035 @default.
- W4313457532 hasConceptScore W4313457532C154945302 @default.
- W4313457532 hasConceptScore W4313457532C169258074 @default.