Detecting Outliers with Poisson Image Interpolation

TY - CHAP

T1 - Detecting Outliers with Poisson Image Interpolation

AU - Tan, Jeremy

AU - Hou, Benjamin

AU - Day, Thomas

AU - Simpson, John

AU - Rueckert, Daniel

AU - Kainz, Bernhard

N1 - Funding Information: Acknowledgements. Support from Wellcome Trust IEH Award iFind project [102431] and UK Research and Innovation London Medical Imaging and Artificial Intelligence Centre for Value Based Healthcare. JT was supported by the ICL President’s Scholarship. Publisher Copyright: © 2021, Springer Nature Switzerland AG.

PY - 2021

Y1 - 2021

N2 - Supervised learning of every possible pathology is unrealistic for many primary care applications like health screening. Image anomaly detection methods that learn normal appearance from only healthy data have shown promising results recently. We propose an alternative to image reconstruction-based and image embedding-based methods and propose a new self-supervised method to tackle pathological anomaly detection. Our approach originates in the foreign patch interpolation (FPI) strategy that has shown superior performance on brain MRI and abdominal CT data. We propose to use a better patch interpolation strategy, Poisson image interpolation (PII), which makes our method suitable for applications in challenging data regimes. PII outperforms state-of-the-art methods by a good margin when tested on surrogate tasks like identifying common lung anomalies in chest X-rays or hypo-plastic left heart syndrome in prenatal, fetal cardiac ultrasound images. Code available at https://github.com/jemtan/PII.

AB - Supervised learning of every possible pathology is unrealistic for many primary care applications like health screening. Image anomaly detection methods that learn normal appearance from only healthy data have shown promising results recently. We propose an alternative to image reconstruction-based and image embedding-based methods and propose a new self-supervised method to tackle pathological anomaly detection. Our approach originates in the foreign patch interpolation (FPI) strategy that has shown superior performance on brain MRI and abdominal CT data. We propose to use a better patch interpolation strategy, Poisson image interpolation (PII), which makes our method suitable for applications in challenging data regimes. PII outperforms state-of-the-art methods by a good margin when tested on surrogate tasks like identifying common lung anomalies in chest X-rays or hypo-plastic left heart syndrome in prenatal, fetal cardiac ultrasound images. Code available at https://github.com/jemtan/PII.

KW - Outlier detection

KW - Self-supervised learning

UR - http://www.scopus.com/inward/record.url?scp=85115388153&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-87240-3_56

DO - 10.1007/978-3-030-87240-3_56

M3 - Conference paper

AN - SCOPUS:85115388153

SN - 9783030872397

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 581

EP - 591

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 - 24th International Conference, Proceedings

A2 - de Bruijne, Marleen

A2 - Cattin, Philippe C.

A2 - Cotin, Stéphane

A2 - Padoy, Nicolas

A2 - Speidel, Stefanie

A2 - Zheng, Yefeng

A2 - Essert, Caroline

PB - Springer Science and Business Media Deutschland GmbH

T2 - 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021

Y2 - 27 September 2021 through 1 October 2021

ER -