Producing radiologist quality reports for interpretable deep learning

Simple item page
dc.contributor.authorGale, W.
dc.contributor.authorOakden-Rayner, L.
dc.contributor.authorCarneiro, G.
dc.contributor.authorPalmer, L.J.
dc.contributor.authorBradley, A.P.
dc.contributor.conferenceIEEE International Symposium on Biomedical Imaging (ISBI) (8 Apr 2019 - 11 Apr 2019 : Venice, ITALY)
dc.date.issued2019
dc.description.abstractCurrent approaches to explaining the decisions of deep learning systems for medical tasks have focused on visualising the elements that have contributed to each decision. We argue that such approaches are not enough to “open the black box” of medical decision making systems because they are missing a key component that has been used as a standard communication tool between doctors for centuries: language. We propose a model-agnostic interpretability method that involves training a simple recurrent neural network model to produce descriptive sentences to clarify the decision of deep learning classifiers. We test our method on the task of detecting hip fractures from frontal pelvic x-rays. This process requires minimal additional labelling despite producing text containing elements that the original deep learning classification model was not specifically trained to detect. The experimental results show that: 1) the sentences produced by our method consistently contain the desired information, 2) the generated sentences are preferred by the cohort of doctors tested compared to current tools that create saliency maps, and 3) the combination of visualisations and generated text is better than either alone.
dc.description.statementofresponsibilityWilliam Gale, Luke Oakden-Rayner, Gustavo Carneiro, Lyle J. Palmer, Andrew P. Bradley
dc.identifier.citationProceedings / IEEE International Symposium on Biomedical Imaging: from nano to macro. IEEE International Symposium on Biomedical Imaging, 2019, vol.2019-April, pp.1275-1279
dc.identifier.doi10.1109/ISBI.2019.8759236
dc.identifier.isbn9781538636411
dc.identifier.issn1945-7928
dc.identifier.issn1945-8452
dc.identifier.orcidOakden-Rayner, L. [0000-0001-5471-5202]
dc.identifier.orcidCarneiro, G. [0000-0002-5571-6220]
dc.identifier.orcidPalmer, L.J. [0000-0002-1628-3055]
dc.identifier.urihttp://hdl.handle.net/2440/121648
dc.language.isoen
dc.publisherIEEE
dc.publisher.placeonline
dc.relation.granthttp://purl.org/au-research/grants/arc/DP180103232
dc.relation.ispartofseriesIEEE International Symposium on Biomedical Imaging
dc.rights© 2019 IEEE
dc.source.urihttps://doi.org/10.1109/isbi.2019.8759236
dc.subjectPattern recognition; text generation; x-ray imaging; bone, fractures
dc.titleProducing radiologist quality reports for interpretable deep learning
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications