FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387054695> ?p ?o ?g. }

• W4387054695 endingPage "662" @default.
• W4387054695 startingPage "662" @default.
• W4387054695 abstract "PURPOSE: The objective of this study was to determine the capability of artificial intelligence (AI) to identify expiratory flow-limitation (EFL) during exercise in adults with varying levels of baseline airway function. We hypothesized that an artificial neural network would accurately identify spontaneous exercise tidal flow-volume loops (eFVL) that exceed maximal airflow from the maximal forced expiration (i.e., EFL). METHODS: Ten adults (6 M, 4 F; mean age = 30.1 yrs) with varying baseline airway caliber completed a graded exercise test-to-exhaustion. At each exercise workload, spontaneous eFVLs were placed within each subject’s pre-exercise maximal expiratory flow-volume curve. Inspiratory capacity maneuvers were performed during exercise for placement of the eFVLs. Each eFVL was coded as flow-limited (1) or non-flow-limited (0). An artificial neural network with three hidden layers and a 2-neuron softmax output layer was used to analyze the eFVLs. The model was trained for 500 epochs. 706 out of a total of 1055 eFVLs (67%) were randomly selected to train the AI model. The remaining 349 eFVLs (33%) were used to test the model's accuracy. RESULTS: Subjects consisted of healthy adults with normal spirometry and asthmatics with mild airway obstruction; forced expiratory volume in 1 second/forced vital capacity averaged 0.7 (range = 0.56 to 0.85). 701 out of the total 1055 eFVLs (66.4%) and 227 out of the 349 eFVLs (65%) used to test the model were flow limited. Following training, the model accurately predicted 267 of the 349 eFVLs in the test data set as flow limited or not flow limited, or 76.5% accuracy. Model sensitivity was 86.3% with 196 eFVLs as true positives (TP) and 31 eFVLs as false negatives (FN). Specificity was 58.3%, with 71 eFVLs as true negatives (TN) and 51 eFVLs as false positives (FP). Precision was determined to be 79.3% (196 TP and 51 FP). CONCLUSION: To our knowledge, this is the first study to explore the potential of AI for identifying expiratory flow limitation in exercising adults. The artificial neural network’s relatively high accuracy and sensitivity suggest that AI could be a viable tool for identifying ventilatory constraint during exercise. A well-developed AI system may be used to streamline the tedious process of identifying EFL in the healthcare setting." @default.
• W4387054695 created "2023-09-27" @default.
• W4387054695 creator A5005043845 @default.
• W4387054695 creator A5005906082 @default.
• W4387054695 creator A5041073385 @default.
• W4387054695 date "2023-09-01" @default.
• W4387054695 modified "2023-09-27" @default.
• W4387054695 title "Artificial Intelligence Determination Of Expiratory Flow Limitation During Exercise In Adults" @default.
• W4387054695 doi "https://doi.org/10.1249/01.mss.0000985988.19808.97" @default.
• W4387054695 hasPublicationYear "2023" @default.
• W4387054695 type Work @default.
• W4387054695 citedByCount "0" @default.
• W4387054695 crossrefType "journal-article" @default.
• W4387054695 hasAuthorship W4387054695A5005043845 @default.
• W4387054695 hasAuthorship W4387054695A5005906082 @default.
• W4387054695 hasAuthorship W4387054695A5041073385 @default.
• W4387054695 hasBestOaLocation W43870546951 @default.
• W4387054695 hasConcept C111919701 @default.
• W4387054695 hasConcept C126322002 @default.
• W4387054695 hasConcept C154945302 @default.
• W4387054695 hasConcept C164705383 @default.
• W4387054695 hasConcept C165637977 @default.
• W4387054695 hasConcept C188441871 @default.
• W4387054695 hasConcept C2776042228 @default.
• W4387054695 hasConcept C2777714996 @default.
• W4387054695 hasConcept C2778476105 @default.
• W4387054695 hasConcept C2780333948 @default.
• W4387054695 hasConcept C3018587741 @default.
• W4387054695 hasConcept C41008148 @default.
• W4387054695 hasConcept C50644808 @default.
• W4387054695 hasConcept C55520419 @default.
• W4387054695 hasConcept C71924100 @default.
• W4387054695 hasConceptScore W4387054695C111919701 @default.
• W4387054695 hasConceptScore W4387054695C126322002 @default.
• W4387054695 hasConceptScore W4387054695C154945302 @default.
• W4387054695 hasConceptScore W4387054695C164705383 @default.
• W4387054695 hasConceptScore W4387054695C165637977 @default.
• W4387054695 hasConceptScore W4387054695C188441871 @default.
• W4387054695 hasConceptScore W4387054695C2776042228 @default.
• W4387054695 hasConceptScore W4387054695C2777714996 @default.
• W4387054695 hasConceptScore W4387054695C2778476105 @default.
• W4387054695 hasConceptScore W4387054695C2780333948 @default.
• W4387054695 hasConceptScore W4387054695C3018587741 @default.
• W4387054695 hasConceptScore W4387054695C41008148 @default.
• W4387054695 hasConceptScore W4387054695C50644808 @default.
• W4387054695 hasConceptScore W4387054695C55520419 @default.
• W4387054695 hasConceptScore W4387054695C71924100 @default.
• W4387054695 hasIssue "9S" @default.
• W4387054695 hasLocation W43870546951 @default.
• W4387054695 hasOpenAccess W4387054695 @default.
• W4387054695 hasPrimaryLocation W43870546951 @default.
• W4387054695 hasRelatedWork W153523923 @default.
• W4387054695 hasRelatedWork W2082677861 @default.
• W4387054695 hasRelatedWork W2140798679 @default.
• W4387054695 hasRelatedWork W2375063122 @default.
• W4387054695 hasRelatedWork W2485672525 @default.
• W4387054695 hasRelatedWork W2887006064 @default.
• W4387054695 hasRelatedWork W2936761846 @default.
• W4387054695 hasRelatedWork W4214510678 @default.
• W4387054695 hasRelatedWork W4289352094 @default.
• W4387054695 hasRelatedWork W4312304745 @default.
• W4387054695 hasVolume "55" @default.
• W4387054695 isParatext "false" @default.
• W4387054695 isRetracted "false" @default.
• W4387054695 workType "article" @default.