HeMIS: Hetero-Modal Image Segmentation

TL;DR

HeMIS is a robust deep learning framework for image segmentation that effectively handles missing modalities by learning a shared latent space, enabling accurate segmentation even with incomplete data.

Contribution

The paper introduces a novel deep learning approach that avoids data imputation by embedding modalities into a common space, allowing flexible and robust segmentation with any subset of available modalities.

Findings

Achieves state-of-the-art results with all modalities available.

Degrades gracefully when modalities are missing, outperforming other synthesis methods.

Effective on neurological MRI datasets for brain tumors and MS lesions.

Abstract

We introduce a deep learning image segmentation framework that is extremely robust to missing imaging modalities. Instead of attempting to impute or synthesize missing data, the proposed approach learns, for each modality, an embedding of the input image into a single latent vector space for which arithmetic operations (such as taking the mean) are well defined. Points in that space, which are averaged over modalities available at inference time, can then be further processed to yield the desired segmentation. As such, any combinatorial subset of available modalities can be provided as input, without having to learn a combinatorial number of imputation models. Evaluated on two neurological MRI datasets (brain tumors and MS lesions), the approach yields state-of-the-art segmentation results when provided with all modalities; moreover, its performance degrades remarkably gracefully when modalities are removed, significantly more so than alternative mean-filling or other synthesis approaches.