Investigating image registration impact on preterm birth classification: An interpretable deep learning approach

Irina Grigorescu*, Lucilio Cordero-Grande, A. David Edwards, Joseph V. Hajnal, Marc Modat, Maria Deprez

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

3 Citations (Scopus)

Abstract

Deep learning algorithms have recently become the dominant trend in medical image classification. However, the decision-making rationale of convolutional neural network (CNN) classifiers can be obscure. Interpretable machine learning techniques, such as layer-wise relevance propagation (LRP), can provide a visual interpretation of these decisions. In this work, we build a 3D CNN model to classify neonatal $$T:2$$ -weighted magnetic resonance (MR) scans into term or preterm. Additionally, we investigate the impact of different registration techniques applied to the image dataset on the classifier’s predictions. Finally, we compute LRP ‘relevance maps’, which indicate each voxel’s importance to the outcome of the decision. Our resulting LRP heatmaps show no visually striking differences between the different registration techniques, while also revealing anatomically plausible features for term and preterm birth.

Original languageEnglish
Title of host publicationSmart Ultrasound Imaging and Perinatal, Preterm and Paediatric Image Analysis - 1st International Workshop, SUSI 2019, and 4th International Workshop, PIPPI 2019, Held in Conjunction with MICCAI 2019, Proceedings
EditorsQian Wang, Alberto Gomez, Jana Hutter, Alberto Gomez, Veronika Zimmer, Jana Hutter, Emma Robinson, Daan Christiaens, Andrew Melbourne, Kristin McLeod, Oliver Zettinig, Roxane Licandro, Esra Abaci Turk
PublisherSPRINGER
Pages104-112
Number of pages9
ISBN (Electronic)9783030328757
ISBN (Print)9783030328740
DOIs
Publication statusPublished - 2019
Event1st International Workshop on Smart Ultrasound Imaging, SUSI 2019, and the 4th International Workshop on Preterm, Perinatal and Paediatric Image Analysis, PIPPI 2019, held in conjunction with the 22nd International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: 17 Oct 201917 Oct 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11798 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st International Workshop on Smart Ultrasound Imaging, SUSI 2019, and the 4th International Workshop on Preterm, Perinatal and Paediatric Image Analysis, PIPPI 2019, held in conjunction with the 22nd International Conference on Medical Imaging and Computer-Assisted Intervention, MICCAI 2019
Country/TerritoryChina
CityShenzhen
Period17/10/201917/10/2019

Keywords

  • Classification
  • Layer-wise relevance propagation
  • Preterm birth

Fingerprint

Dive into the research topics of 'Investigating image registration impact on preterm birth classification: An interpretable deep learning approach'. Together they form a unique fingerprint.

Cite this