Highly accurate classification of chest radiographic reports using a deep learning natural language model pre-trained on 3.8 million text reports.
Keno K. BressemLisa C. AdamsRobert A. GaudinDaniel TröltzschBernd HammMarcus R. MakowskiChan-Yong SchüleJanis Lucas VahldiekStefan Markus NiehuesPublished in: Bioinform. (2021)
Keyphrases
- highly accurate
- language model
- deep learning
- information retrieval
- probabilistic model
- unsupervised learning
- machine learning
- n gram
- high quality
- pre trained
- high accuracy
- pattern recognition
- text mining
- feature selection
- image classification
- speech recognition
- training set
- support vector
- training data
- data mining
- statistical model
- reinforcement learning
- training samples
- decision trees
- model selection
- image segmentation
- feature extraction
- feature vectors