Extremely imbalanced subarachnoid hemorrhage detection based on DenseNet-LSTM network with class-balanced loss and transfer learning.
Zhongyang LuMasahiro OdaYuichiro HayashiTao HuHayato ItohTakeyuki WatadaniOsamu AbeMasahiro HashimotoMasahiro JinzakiKensaku MoriPublished in: Medical Imaging: Computer-Aided Diagnosis (2021)
Keyphrases
- transfer learning
- knowledge transfer
- target class
- prediction with expert advice
- learning tasks
- transfer knowledge
- labeled data
- cross domain
- active learning
- machine learning
- reinforcement learning
- domain adaptation
- semi supervised learning
- text classification
- label information
- machine learning algorithms
- manifold alignment
- imbalanced datasets
- recurrent neural networks
- class labels
- collaborative filtering
- target domain
- feature selection
- multi task learning
- minority class
- information retrieval
- bayesian networks
- support vector machine
- data analysis