Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367396075> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4367396075 endingPage "111606" @default.
- W4367396075 startingPage "111606" @default.
- W4367396075 abstract "Clinical datasets often comprise multiple data points or trials sampled from a single participant. When these datasets are used to train machine learning models, the method used to extract train and test sets must be carefully chosen. Using the standard machine learning approach (random-wise split), different trials from the same participant may appear in both training and test sets. This has led to schemes capable of segregating data points from a same participant into a single set (subject-wise split). Past investigations have demonstrated that models trained in this manner underperform compared to those trained using random-split schemes. Additional training of models via a small subset of trials, known as calibration, bridges the gap in performance across split schemes; however, the amount of calibration trials required to achieve strong model performance is unclear. Thus, this study aims to investigate the relationship between calibration training set size and prediction accuracy on the calibration test set. A database of 30 young, healthy adults performing multiple walking trials across nine different surfaces while fit with inertial measurement unit sensors on the lower limbs was used to develop a deep-learning classifier. For subject-wise trained models, calibration on a single gait cycle per surface yielded a 70% increase in F1-score, the harmonic mean of precision and recall, while 10 gait cycles per surface were sufficient to match the performance of a random-wise trained model. Code to generate calibration curves may be found at (https://github.com/GuillaumeLam/PaCalC)." @default.
- W4367396075 created "2023-04-30" @default.
- W4367396075 creator A5033149401 @default.
- W4367396075 creator A5045494030 @default.
- W4367396075 creator A5055430458 @default.
- W4367396075 date "2023-06-01" @default.
- W4367396075 modified "2023-10-04" @default.
- W4367396075 title "Estimating individual minimum calibration for deep-learning with predictive performance recovery: An example case of gait surface classification from wearable sensor gait data" @default.
- W4367396075 cites W2022145583 @default.
- W4367396075 cites W2068654413 @default.
- W4367396075 cites W2560647685 @default.
- W4367396075 cites W2569508898 @default.
- W4367396075 cites W2595557058 @default.
- W4367396075 cites W2596627175 @default.
- W4367396075 cites W2604834158 @default.
- W4367396075 cites W2610332124 @default.
- W4367396075 cites W2939354476 @default.
- W4367396075 cites W2949497754 @default.
- W4367396075 cites W2951934944 @default.
- W4367396075 cites W3040736388 @default.
- W4367396075 cites W3126854045 @default.
- W4367396075 cites W4225249857 @default.
- W4367396075 cites W4281571001 @default.
- W4367396075 cites W4296425714 @default.
- W4367396075 doi "https://doi.org/10.1016/j.jbiomech.2023.111606" @default.
- W4367396075 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37187130" @default.
- W4367396075 hasPublicationYear "2023" @default.
- W4367396075 type Work @default.
- W4367396075 citedByCount "0" @default.
- W4367396075 crossrefType "journal-article" @default.
- W4367396075 hasAuthorship W4367396075A5033149401 @default.
- W4367396075 hasAuthorship W4367396075A5045494030 @default.
- W4367396075 hasAuthorship W4367396075A5055430458 @default.
- W4367396075 hasConcept C105795698 @default.
- W4367396075 hasConcept C119857082 @default.
- W4367396075 hasConcept C149635348 @default.
- W4367396075 hasConcept C150594956 @default.
- W4367396075 hasConcept C151800584 @default.
- W4367396075 hasConcept C153180895 @default.
- W4367396075 hasConcept C154945302 @default.
- W4367396075 hasConcept C165838908 @default.
- W4367396075 hasConcept C16910744 @default.
- W4367396075 hasConcept C169258074 @default.
- W4367396075 hasConcept C169903167 @default.
- W4367396075 hasConcept C177264268 @default.
- W4367396075 hasConcept C199360897 @default.
- W4367396075 hasConcept C33923547 @default.
- W4367396075 hasConcept C41008148 @default.
- W4367396075 hasConcept C71924100 @default.
- W4367396075 hasConcept C79061980 @default.
- W4367396075 hasConcept C95623464 @default.
- W4367396075 hasConcept C99508421 @default.
- W4367396075 hasConceptScore W4367396075C105795698 @default.
- W4367396075 hasConceptScore W4367396075C119857082 @default.
- W4367396075 hasConceptScore W4367396075C149635348 @default.
- W4367396075 hasConceptScore W4367396075C150594956 @default.
- W4367396075 hasConceptScore W4367396075C151800584 @default.
- W4367396075 hasConceptScore W4367396075C153180895 @default.
- W4367396075 hasConceptScore W4367396075C154945302 @default.
- W4367396075 hasConceptScore W4367396075C165838908 @default.
- W4367396075 hasConceptScore W4367396075C16910744 @default.
- W4367396075 hasConceptScore W4367396075C169258074 @default.
- W4367396075 hasConceptScore W4367396075C169903167 @default.
- W4367396075 hasConceptScore W4367396075C177264268 @default.
- W4367396075 hasConceptScore W4367396075C199360897 @default.
- W4367396075 hasConceptScore W4367396075C33923547 @default.
- W4367396075 hasConceptScore W4367396075C41008148 @default.
- W4367396075 hasConceptScore W4367396075C71924100 @default.
- W4367396075 hasConceptScore W4367396075C79061980 @default.
- W4367396075 hasConceptScore W4367396075C95623464 @default.
- W4367396075 hasConceptScore W4367396075C99508421 @default.
- W4367396075 hasLocation W43673960751 @default.
- W4367396075 hasLocation W43673960752 @default.
- W4367396075 hasOpenAccess W4367396075 @default.
- W4367396075 hasPrimaryLocation W43673960751 @default.
- W4367396075 hasRelatedWork W2792951589 @default.
- W4367396075 hasRelatedWork W2911455822 @default.
- W4367396075 hasRelatedWork W3103186320 @default.
- W4367396075 hasRelatedWork W3154710845 @default.
- W4367396075 hasRelatedWork W3159716340 @default.
- W4367396075 hasRelatedWork W3207189946 @default.
- W4367396075 hasRelatedWork W4280641190 @default.
- W4367396075 hasRelatedWork W4308191010 @default.
- W4367396075 hasRelatedWork W4323021782 @default.
- W4367396075 hasRelatedWork W4366147362 @default.
- W4367396075 hasVolume "154" @default.
- W4367396075 isParatext "false" @default.
- W4367396075 isRetracted "false" @default.
- W4367396075 workType "article" @default.