Revisiting MAE pre-training for 3D medical image segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wald, Tassilo, Ulrich, Constantin, Lukyanenko, Stanislav, Goncharov, Andrei, Paderno, Alberto, Miller, Maximilian, Maerkisch, Leander, Jäger, Paul F., Maier-Hein, Klaus
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908300481658880
author Wald, Tassilo
Ulrich, Constantin
Lukyanenko, Stanislav
Goncharov, Andrei
Paderno, Alberto
Miller, Maximilian
Maerkisch, Leander
Jäger, Paul F.
Maier-Hein, Klaus
author_facet Wald, Tassilo
Ulrich, Constantin
Lukyanenko, Stanislav
Goncharov, Andrei
Paderno, Alberto
Miller, Maximilian
Maerkisch, Leander
Jäger, Paul F.
Maier-Hein, Klaus
contents Self-Supervised Learning (SSL) presents an exciting opportunity to unlock the potential of vast, untapped clinical datasets, for various downstream applications that suffer from the scarcity of labeled data. While SSL has revolutionized fields like natural language processing and computer vision, its adoption in 3D medical image computing has been limited by three key pitfalls: Small pre-training dataset sizes, architectures inadequate for 3D medical image analysis, and insufficient evaluation practices. In this paper, we address these issues by i) leveraging a large-scale dataset of 39k 3D brain MRI volumes and ii) using a Residual Encoder U-Net architecture within the state-of-the-art nnU-Net framework. iii) A robust development framework, incorporating 5 development and 8 testing brain MRI segmentation datasets, allowed performance-driven design decisions to optimize the simple concept of Masked Auto Encoders (MAEs) for 3D CNNs. The resulting model not only surpasses previous SSL methods but also outperforms the strong nnU-Net baseline by an average of approximately 3 Dice points setting a new state-of-the-art. Our code and models are made available here.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23132
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Revisiting MAE pre-training for 3D medical image segmentation
Wald, Tassilo
Ulrich, Constantin
Lukyanenko, Stanislav
Goncharov, Andrei
Paderno, Alberto
Miller, Maximilian
Maerkisch, Leander
Jäger, Paul F.
Maier-Hein, Klaus
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Self-Supervised Learning (SSL) presents an exciting opportunity to unlock the potential of vast, untapped clinical datasets, for various downstream applications that suffer from the scarcity of labeled data. While SSL has revolutionized fields like natural language processing and computer vision, its adoption in 3D medical image computing has been limited by three key pitfalls: Small pre-training dataset sizes, architectures inadequate for 3D medical image analysis, and insufficient evaluation practices. In this paper, we address these issues by i) leveraging a large-scale dataset of 39k 3D brain MRI volumes and ii) using a Residual Encoder U-Net architecture within the state-of-the-art nnU-Net framework. iii) A robust development framework, incorporating 5 development and 8 testing brain MRI segmentation datasets, allowed performance-driven design decisions to optimize the simple concept of Masked Auto Encoders (MAEs) for 3D CNNs. The resulting model not only surpasses previous SSL methods but also outperforms the strong nnU-Net baseline by an average of approximately 3 Dice points setting a new state-of-the-art. Our code and models are made available here.
title Revisiting MAE pre-training for 3D medical image segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2410.23132