Investigation of the best effective fold of data augmentation for training deep learning models for recognition of contiguity between mandibular third molar and inferior alveolar canal on panoramic radiographs.

Dhanaporn Papasratorn Suchaya Pornprasertsuk-Damrongsri Suraphong Yuma Warangkana Weerawanich

Published in: Clinical oral investigations (2023)

Ten-fold augmentation may help improve deep learning models' performances. The variety of original data and the accuracy of labels are essential to train a high-performance model.

Keyphrases

deep learning
electronic health record
big data
artificial intelligence
convolutional neural network
machine learning
soft tissue
high speed
high resolution
cone beam computed tomography