Abstract:An important challenge and limiting factor in deep learning methods for medical imaging segmentation is the lack of available of annotated data to properly train models. For the specific task of tumor segmentation, the process entails clinicians labeling every slice of volumetric scans for every patient, which becomes prohibitive at the scale of datasets required to train neural networks to optimal performance. To address this, we propose a novel semi-supervised framework that allows training any segmentation (encoder–decoder) model using only information readily available in radiological data, namely the presence of a tumor in the image, in addition to a few annotated images. Specifically, we conjecture that a generative model performing domain translation on this weak label — healthy vs diseased scans — helps achieve tumor segmentation. The proposed GenSeg method first disentangles tumoral tissue from healthy “background” tissue. The latent representation is separated into (1) the common background information across both domains, and (2) the unique tumoral information. GenSeg then achieves diseased-to-healthy image translation by decoding a healthy version of the image from just the common representation, as well as a residual image that allows adding back the tumors. The same decoder that produces this residual tumor image, also outputs a tumor segmentation. Implicit data augmentation is achieved by re-using the same framework for healthy-to-diseased image translation, where a residual tumor image is produced from a prior distribution. By performing both image translation and segmentation simultaneously, GenSeg allows training on only partially annotated datasets. To test the framework, we trained U-Net-like architectures using GenSeg and evaluated their performance on 3 variants of a synthetic task, as well as on 2 benchmark datasets: brain tumor segmentation in MRI (derived from BraTS) and liver metastasis segmentation in CT (derived from LiTS). Our method outperforms the baseline semi-supervised (autoencoder and mean teacher) and supervised segmentation methods, with improvements ranging between 8–14% Dice score on the brain task and 5–8% on the liver task, when only 1% of the training images were annotated. These results show the proposed framework is ideal at addressing the problem of training deep segmentation models when a large portion of the available data is unlabeled and unpaired, a common issue in tumor segmentation.

Annotation-Free Gliomas Segmentation Based on a Few Labeled General Brain Tumor Images

LF-SynthSeg: Label-Free Brain Tissue-Assisted Tumor Synthesis and Segmentation

Towards annotation-efficient segmentation via image-to-image translation

Glioma Segmentation Using a Novel Unified Algorithm in Multimodal MRI Images

Cross-Modality Brain Tumor Segmentation via Bidirectional Global-to-Local Unsupervised Domain Adaptation

An unsupervised domain adaptation brain CT segmentation method across image modalities and diseases

A Comprehensive Multi-modal Domain Adaptative Aid Framework for Brain Tumor Diagnosis.

DG-TTA: Out-of-domain medical image segmentation through Domain Generalization and Test-Time Adaptation

ADAPTIVE TRANSFER LEARNING TO ENHANCE DOMAIN TRANSFER IN BRAIN TUMOR SEGMENTATION

MTANS: Multi-Scale Mean Teacher Combined Adversarial Network with Shape-Aware Embedding for Semi-Supervised Brain Lesion Segmentation

Unsupervised Domain Adaptation for Pediatric Brain Tumor Segmentation

Semi-Supervised Segmentation Framework for Gastrointestinal Lesion Diagnosis in Endoscopic Images

Parasitic GAN for Semi-Supervised Brain Tumor Segmentation.

Unsupervised Domain Adaptation for Brain Structure Segmentation Via Mutual Information Maximization Alignment

TumorGAN: A Multi-Modal Data Augmentation Framework for Brain Tumor Segmentation

Suggestive Annotation of Brain Tumour Images with Gradient-guided Sampling

A General Global and Local Pre-Training Framework for 3D Medical Image Segmentation.

Semi-supervised Gastrointestinal Lesion Segmentation using Adversarial Learning

Motion-artifact-augmented pseudo-label network for semi-supervised brain tumor segmentation

Adult Glioma Segmentation in Sub-Saharan Africa using Transfer Learning on Stratified Finetuning Data

Few Exemplar-Based General Medical Image Segmentation via Domain-Aware Selective Adaptation