DLTK: State of the Art Reference Implementations for Deep Learning on Medical Images

TL;DR

DLTK is a modular TensorFlow-based toolkit that provides state-of-the-art deep learning implementations for medical image analysis, achieving top performance on benchmark segmentation challenges.

Contribution

It introduces a comprehensive, easy-to-use toolkit with reference implementations that outperform previous methods on medical image segmentation benchmarks.

Findings

Achieved a Dice similarity coefficient of 81.5 on a public challenge dataset.

Outperformed previous CNN and challenge-winning methods in segmentation accuracy.

Provided a modular toolkit facilitating rapid experimentation in medical imaging deep learning.

Abstract

We present DLTK, a toolkit providing baseline implementations for efficient experimentation with deep learning methods on biomedical images. It builds on top of TensorFlow and its high modularity and easy-to-use examples allow for a low-threshold access to state-of-the-art implementations for typical medical imaging problems. A comparison of DLTK's reference implementations of popular network architectures for image segmentation demonstrates new top performance on the publicly available challenge data "Multi-Atlas Labeling Beyond the Cranial Vault". The average test Dice similarity coefficient of $81.5$ exceeds the previously best performing CNN ($75.7$) and the accuracy of the challenge winning method ($79.0$).