Publications by authors named "Vincent Tze Yang Tiong"
Rationale And Objectives: To compare the performance of pneumothorax deep learning detection models trained with radiologist versus natural language processing (NLP) labels on the NIH ChestX-ray14 dataset.
Materials And Methods: The ChestX-ray14 dataset consisted of 112,120 frontal chest radiographs with 5302 positive and 106, 818 negative labels for pneumothorax using NLP (dataset A). All 112,120 radiographs were also inspected by 4 radiologists leaving a visually confirmed set of 5,138 positive and 104,751 negative for pneumothorax (dataset B).
Article Synopsis
- * The model was trained on two large datasets and tested across six external datasets, achieving high accuracy (AUC scores ranging from 0.91 to 0.98) in detecting pneumothorax compared to a 0.93 AUC in internal testing.
- * The results indicate that the model performs better in identifying larger pneumothoraces compared to smaller ones, and the presence or absence of a chest tube on radiographs does not significantly affect detection accuracy.
Background: The use of live and cadaveric animal models in surgical training is well established as a means of teaching and improving surgical skill in a controlled setting. We aim to review, evaluate, and summarize the models published in the literature that are applicable to Plastic Surgery training.
Materials And Methods: A PubMed search for keywords relating to animal models in Plastic Surgery and the associated procedures was conducted.