King's College London

Research portal

Learning task-specific and shared representations in medical imaging

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Felix J.S. Bragman, Ryutaro Tanno, Sebastien Ourselin, Daniel C. Alexander, M. Jorge Cardoso

Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
EditorsDinggang Shen, Pew-Thian Yap, Tianming Liu, Terry M. Peters, Ali Khan, Lawrence H. Staib, Caroline Essert, Sean Zhou
Number of pages10
ISBN (Print)9783030322502
Publication statusPublished - 2019
Event22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: 13 Oct 201917 Oct 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11767 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019

King's Authors


The performance of multi-task learning hinges on the design of feature sharing between tasks; a process which is combinatorial in the network depth and task count. Hand-crafting an architecture based on human intuitions of task relationships is therefore suboptimal. In this paper, we present a probabilistic approach to learning task-specific and shared representations in Convolutional Neural Networks (CNNs) for multi-task learning of semantic tasks. We introduce Stochastic Filter Groups; which is a mechanism that groups convolutional kernels into task-specific and shared groups to learn an optimal kernel allocation. They facilitate learning optimal shared and task specific representations. We employ variational inference to learn the posterior distribution over the possible grouping of kernels and CNN weights. Experiments on MRI-based prostate radiotherapy organ segmentation and CT synthesis demonstrate that the proposed method learns optimal task allocations that are inline with human-optimised networks whilst improving performance over competing baselines.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454