MULTIPLE INSTANCE LEARNING FOR HISTOPATHOLOGY CLASSIFICATION

摘要

The present invention relates to a multi-instance learning method for histopathology classification, a multi-instance learning method for histopathology classification performed by at least one processor in a computing device or a computing network, and a feature extraction model (Fθ(·)) to convert the i-th slide-derived instance (pij) to a low-dimensional embedding (gij) by executing An instance selection step of sampling the highest instance per slide for learning by classifying instance level probabilities, and learning using the instances obtained in the instance selection step, but sequentially performing instance level learning and vowel level learning to obtain a final loss It may include a learning step of obtaining , and a soft assignment-based reasoning step of distributing vowel level embeddings (zi) to learned centroids using a kernel that detects the similarity between two points.