Towards Explainable End-to-End Prostate Cancer Relapse Prediction from H&E Images Combining Self-Attention Multiple Instance Learning with a Recurrent Neural Network.
Esther DietrichPatrick FuhlertAnne ErnstGuido SauterMaximilian LennartzH. Siegfried StiehlMarina ZimmermannStefan BonnPublished in: ML4H@NeurIPS (2021)
Keyphrases
- end to end
- multiple instance learning
- image categorization
- object based image retrieval
- prostate cancer
- image annotation
- multiple instance
- text localization and recognition
- prostate segmentation
- image analysis
- image data
- image database
- multi class
- image set
- input image
- image features
- image understanding
- image retrieval
- supervised learning
- computer aided
- similarity measure
- semi supervised learning
- image registration
- mr images
- image regions
- visual features
- multi modal
- image classification
- support vector machine
- data points
- object recognition
- pattern recognition
- decision trees