Open Access Green: Postprint online verfügbar 11/2020 möglich sobald bei der ZB eingereicht worden ist.
Multi-task learning of a deep K-nearest neighbour network for histopathological image classification and retrieval.
Lecture Notes Comp. Sci. 11764 LNCS, 676-684 (2019)
DOI Verlagsversion bestellen
Deep neural networks have achieved tremendous success in image recognition, classification and object detection. However, deep learning is often criticised for its lack of transparency and general inability to rationalise its predictions. The issue of poor model interpretability becomes critical in medical applications: a model that is not understood and trusted by physicians is unlikely to be used in daily clinical practice. In this work, we develop a novel multi-task deep learning framework for simultaneous histopathology image classification and retrieval, leveraging on the classic concept of k-nearest neighbours to improve model interpretability. For a test image, we retrieve the most similar images from our training databases. These retrieved nearest neighbours can be used to classify the test image with a confidence score, and provide a human-interpretable explanation of our classification. Our original framework can be built on top of any existing classification network (and therefore benefit from pretrained models), by (i) combining a triplet loss function with a novel triplet sampling strategy to compare distances between samples and (ii) adding a Cauchy hashing loss function to accelerate neighbour searching. We evaluate our method on colorectal cancer histology slides and show that the confidence estimates are strongly correlated with model performance. Nearest neighbours are intuitive and useful for expert evaluation. They give insights into understanding possible model failures, and can support clinical decision making by comparing archived images and patient records with the actual case.
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Journalartikel
Dokumenttyp Wissenschaftlicher Artikel
ISSN (print) / ISBN 0302-9743
Zeitschrift Lecture Notes in Computer Science
Quellenangaben Band: 11764 LNCS, Seiten: 676-684
Verlagsort Berlin [u.a.]
Institut(e) Institute of Computational Biology (ICB)