MixUp-MIL: Novel Data Augmentation for Multiple Instance Learning and a Study on Thyroid Cancer Diagnosis

Michael Gadermayr* (First author), Lukas Koller, Maximilian Tschuchnig, Lea Maria Stangassigner, Christina Erhardt-Kreutzer (Co-author), Sébastien Couillard-Després (Co-author), Gertie Janneke Oostingh, Anton Hittmair

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

Multiple instance learning is a powerful approach for whole slide image-based diagnosis in the absence of pixel- or patch-level annotations. In spite of the huge size of whole slide images, the number of individual slides is often rather small, leading to a small number of labeled samples. To improve training, we propose and investigate novel data augmentation strategies for multiple instance learning based on the idea of linear and multilinear interpolation of feature vectors within and between individual whole slide images. Based on state-of-the-art multiple instance learning architectures and two thyroid cancer data sets, an exhaustive study was conducted considering a range of common data augmentation strategies. Whereas a strategy based on to the original MixUp approach showed decreases in accuracy, a novel multilinear intra-slide interpolation method led to consistent increases in accuracy.
Original languageEnglish
Title of host publicationMedical Image Computing and Computer Assisted Intervention – MICCAI 2023
Pages477-486
ISBN (Electronic)978-3-031-43987-2
DOIs
Publication statusPublished - 2023
Event26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) - Vancouver, Canada
Duration: 8 Oct 202312 Oct 2023

Publication series

NameLecture Notes in Computer Science (LNCS)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference with scientific content

Conference with scientific content26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)
Country/TerritoryCanada
CityVancouver
Period8/10/2312/10/23

Fingerprint

Dive into the research topics of 'MixUp-MIL: Novel Data Augmentation for Multiple Instance Learning and a Study on Thyroid Cancer Diagnosis'. Together they form a unique fingerprint.

Cite this