Zhang, Qiang, Bhalerao, Abhir and Hutchinson, Charles E. (2017) Weakly-supervised evidence pinpointing and description. In: Niethammer, M., (ed.) Information Processing in Medical Imaging. IPMI 2017. Lecture Notes in Computer Science, 10265 . Cham: Springer, pp. 210-222. ISBN 9783319590493

Preview

PDF WRAP-weakly-supervised-evidence-pinpointing-Bhalerao-2017 description-.pdf - Accepted Version - Requires a PDF viewer. Download (4Mb) | Preview

Request Changes to record.

Abstract

We propose a learning method to identify which specific regions and features of images contribute to a certain classification. In the medical imaging context, they can be the evidence regions where the abnormalities are most likely to appear, and the discriminative features of these regions supporting the pathology classification. The learning is weakly-supervised requiring only the pathological labels and no other prior knowledge. The method can also be applied to learn the salient description of an anatomy discriminative from its background, in order to localise the anatomy before a classification step. We formulate evidence pinpointing as a sparse descriptor learning problem. Because of the large computational complexity, the objective function is composed in a stochastic way and is optimised by the Regularised Dual Averaging algorithm. We demonstrate that the learnt feature descriptors contain more specific and better discriminative information than hand-crafted descriptors contributing to superior performance for the tasks of anatomy localisation and pathology classification respectively. We apply our method on the problem of lumbar spinal stenosis for localising and classifying vertebrae in MRI images. Experimental results show that our method when trained with only target labels achieves better or competitive performance on both tasks compared with strongly-supervised methods requiring labels and multiple landmarks. A further improvement is achieved with training on additional weakly annotated data, which gives robust localisation with average error within 2 mm and classification accuracies close to human performance.

Item Type:	Book Item
Subjects:	Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software R Medicine > RD Surgery
Divisions:	Faculty of Science, Engineering and Medicine > Science > Computer Science Faculty of Science, Engineering and Medicine > Medicine > Warwick Medical School
Library of Congress Subject Headings (LCSH):	Magnetic resonance imaging, Spinal canal -- Stenosis -- Diagnosis
Series Name:	Lecture Notes in Computer Science
Journal or Publication Title:	Information Processing in Medical Imaging 2017
Publisher:	Springer
Place of Publication:	Cham
ISBN:	9783319590493
Book Title:	Information Processing in Medical Imaging. IPMI 2017
Editor:	Niethammer, M.
Official Date:	9 February 2017
Dates:	Date Event 9 February 2017 Accepted
Volume:	10265
Page Range:	pp. 210-222
DOI:	10.1007/978-3-319-59050-9_17
Status:	Peer Reviewed
Publication Status:	Published
Access rights to Published version:	Restricted or Subscription Access
Date of first compliant deposit:	2 March 2017
Date of first compliant Open Access:	2 March 2017
Conference Paper Type:	Paper
Title of Event:	Information Processing in Medical Imaging 2017
Type of Event:	Conference
Location of Event:	Boone, North Carolina, USA
Date(s) of Event:	25-30 Jun 2017
Related URLs:	Organisation

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads