Unsupervised brain tumor segmentation is pivotal in realms of disease diagnosis, surgical planning, and treatment response monitoring, with the distinct advantage of obviating the need for labeled data. Traditional methodologies in this domain, however, often fall short in fully capitalizing on the extensive prior knowledge of brain tissue, typically approaching the task merely as an anomaly detection challenge. In our research, we present an innovative strategy that effectively integrates brain tissues' prior knowledge into both the synthesis and segmentation of brain tumor from T2-weighted Magnetic Resonance Imaging scans. Central to our method is the tumor synthesis mechanism, employing randomly generated ellipsoids in conjunction with the intensity profiles of brain tissues. This methodology not only fosters a significant degree of variation in the tumor presentations within the synthesized images but also facilitates the creation of an essentially unlimited pool of abnormal T2-weighted images. These synthetic images closely replicate the characteristics of real tumor-bearing scans. Our training protocol extends beyond mere tumor segmentation; it also encompasses the segmentation of brain tissues, thereby directing the network's attention to the boundary relationship between brain tumor and brain tissue, thus improving the robustness of our method. We evaluate our approach across five widely recognized public datasets (BRATS 2019, BRATS 2020, BRATS 2021, PED and SSA), and the results show that our method outperforms state-of-the-art unsupervised tumor segmentation methods by large margins. Moreover, the proposed method achieves more than 92 of the fully supervised performance on the same testing datasets.

Self-supervised learning (SSL) has been proposed to alleviate neural networks' reliance on annotated data and to improve downstream tasks' performance, which has obtained substantial success in several volumetric medical image segmentation tasks. However, most existing approaches are designed and pre-trained on CT or MRI datasets of non-brain organs. The lack of brain prior limits those methods' performance on brain segmentation, especially on fine-grained brain parcellation. To overcome this limitation, we here propose a novel SSL strategy for MRI of the human brain, named Masked Deformation Modeling (MDM). MDM first conducts atlas-guided patch sampling on individual brain MRI scans (moving volumes) and an MNI152 template (a fixed volume). The sampled moving volumes are randomly masked in a feature-aligned manner, and then sent into a U-Net-based network to extract latent features. An intensity head and a deformation field head are used to decode the latent features, respectively restoring the masked volume and predicting the deformation field from the moving volume to the fixed volume. The proposed MDM is fine-tuned and evaluated on three brain parcellation datasets with different granularities (JHU, Mindboggle-101, CANDI), a brain lesion segmentation dataset (ATLAS2), and a brain tumor segmentation dataset (BraTS21). Results demonstrate that MDM outperforms various state-of-the-art medical SSL methods by considerable margins, and can effectively reduce the annotation effort by at least 40%. Codes and pre-trained weights will be released at https://github.com/CRazorback/MDM.

Self-supervised pre-training techniques based on image reconstruction have achieved substantial success in medical image analysis, allowing for the transferability of pre-trained model weights to various downstream tasks for further fine-tuning. However, current pre-training methods primarily target single-modal medical images, like CT scans, scarcely considering the multi-modal images like multi-modal brain MRI. Yet, the latter is especially pivotal for accurate tumor segmentation, given that each modality provides unique insights into the tumor’s characteristics. In this study, we introduce a self-supervised pre-training approach tailored for multi-modal brain MRI, equipped with Masked Modality Complementary Modeling (MMCM). Specifically, the proposed method involves masking a designated portion of each modality to ensure the visible parts are distinct and complementary. We assume that through the process of reconstructing such images, the model not only learns general anatomical and modality-specific characteristics but also gains insights into bridging and mapping across various modalities. Results of downstream experiments validate that our method outperforms state-of-the-art self-supervised learning methods in the tumor segmentation accuracy on the BraTS 2021 dataset. Additionally, in downstream tasks in two common scenarios: the small sample size and the single available modality, our method substantially improves the performance over the baseline model trained from scratch. The model and code are available at https://github.com/liangjianwen01/MMCM.

Due to time and cost limitations, Magnetic Resonance (MR) imaging often employs anisotropic scanning with large slice spacing and thickness. This causes blurring in views perpendicular to the slices, which adversely affects clinical diagnosis and research. Taking into account the complementary information from the reference modality, deep learning (DL) based multi-contrast methods have become a focal point of research. These methods aim to reconstruct the isotropic target MR image with the auxiliary high-resolution (HR) reference modality. However, most of the methods primarily concentrate on the structural restoration of the target low-resolution (LR) image, neglecting the crucial aspect that the coexisting structural and modality differences between target and reference modalities can impede effective restoration. Additionally, these methods are designed for a fixed upsampling scale, not accounting for the practical scenario of varying slice thickness. In this work, we propose a joint Super-resolution and Modality translation network (SMNet) for multi-contrast arbitrary-scale isotropic MRI reconstruction. The modality translation branch includes the Modality-Specific-Augmented Alignment (MSAA) block, which eliminates modality distribution disparities and enhances modality-specific regions on the reference feature before fusion. And the super-resolution branch employs the Reliability-based Spatial Fusion (RSF) block for the structural restoration of the target LR feature using a reliability prior. The outputs from these two branches are then ensembled to obtain the final reconstructed result. Extensive experiments on both a private dataset and the Brasts2021 dataset demonstrate the effectiveness and generalizability of the proposed method. Our code is available at https://github.com/11710615/smnet.

Self-supervised Learning (SSL) has been widely applied to learn image representations through exploiting unlabeled images. However, it has not been fully explored in the medical image analysis field. In this work, Saliency-guided Self-Supervised image Transformer (SSiT) is proposed for Diabetic Retinopathy (DR) grading from fundus images. We novelly introduce saliency maps into SSL, with a goal of guiding self-supervised pre-training with domain-specific prior knowledge. Specifically, two saliency-guided learning tasks are employed in SSiT: 1) Saliency-guided contrastive learning is conducted based on the momentum contrast, wherein fundus images' saliency maps are utilized to remove trivial patches from the input sequences of the momentum-updated key encoder. Thus, the key encoder is constrained to provide target representations focusing on salient regions, guiding the query encoder to capture salient features. 2) The query encoder is trained to predict the saliency segmentation, encouraging the preservation of fine-grained information in the learned representations. To assess our proposed method, four publicly-accessible fundus image datasets are adopted. One dataset is employed for pre-training, while the three others are used to evaluate the pre-trained models' performance on downstream DR grading. The proposed SSiT significantly outperforms other representative state-of-the-art SSL methods on all downstream datasets and under various evaluation settings. For example, SSiT achieves a Kappa score of 81.88% on the DDR dataset under fine-tuning evaluation, outperforming all other ViT-based SSL methods by at least 9.48%.

Contrastive learning based vision-language joint pre-training has emerged as a successful representation learning strategy. In this paper, we present a prototype representation learning framework incorporating both global and local alignment between medical images and reports. In contrast to standard global multi-modality alignment methods, we employ a local alignment module for fine-grained representation. Furthermore, a cross-modality conditional reconstruction module is designed to interchange information across modalities in the training phase by reconstructing masked images and reports. For reconstructing long reports, a sentence-wise prototype memory bank is constructed, enabling the network to focus on low-level localized visual and high-level clinical linguistic features. Additionally, a non-auto-regressive generation paradigm is proposed for reconstructing non-sequential reports. Experimental results on five downstream tasks, including supervised classification, zero-shot classification, image-to-text retrieval, semantic segmentation, and object detection, show the proposed method outperforms other state-of-the-art methods across multiple datasets and under different dataset size settings. The code is available at https://github.com/QtacierP/PRIOR.

Skin lesion segmentation from dermoscopic images plays a vital role in early diagnoses and prognoses of various skin diseases. However, it is a challenging task due to the large variability of skin lesions and their blurry boundaries. Moreover, most existing skin lesion datasets are designed for disease classification, with relatively fewer segmentation labels having been provided. To address these issues, we propose a novel automatic superpixel-based masked image modeling method, named autoSMIM, in a self-supervised setting for skin lesion segmentation. It explores implicit image features from abundant unlabeled dermoscopic images. autoSMIM begins with restoring an input image with randomly masked superpixels. The policy of generating and masking superpixels is then updated via a novel proxy task through Bayesian Optimization. The optimal policy is subsequently used for training a new masked image modeling model. Finally, we finetune such a model on the downstream skin lesion segmentation task. Extensive experiments are conducted on three skin lesion segmentation datasets, including ISIC 2016, ISIC 2017, and ISIC 2018. Ablation studies demonstrate the effectiveness of superpixel-based masked image modeling and establish the adaptability of autoSMIM. Comparisons with state-of-the-art methods show the superiority of our proposed autoSMIM. The source code is available at https://github.com/Wzhjerry/autoSMIM.

Convolutional neural networks have been widely applied to medical image segmentation and have achieved considerable performance. However, the performance may be significantly affected by the domain gap between training data (source domain) and testing data (target domain). To address this issue, we propose a data manipulation based domain generalization method, called Automated Augmentation for Domain Generalization (AADG). Our AADG framework can effectively sample data augmentation policies that generate novel domains and diversify the training set from an appropriate search space. Specifically, we introduce a novel proxy task maximizing the diversity among multiple augmented novel domains as measured by the Sinkhorn distance in a unit sphere space, making automated augmentation tractable. Adversarial training and deep reinforcement learning are employed to efficiently search the objectives. Quantitative and qualitative experiments on 11 publicly-accessible fundus image datasets (four for retinal vessel segmentation, four for optic disc and cup (OD/OC) segmentation and three for retinal lesion segmentation) are comprehensively performed. Two OCTA datasets for retinal vasculature segmentation are further involved to validate cross-modality generalization. Our proposed AADG exhibits state-of-the-art generalization performance and outperforms existing approaches by considerable margins on retinal vessel, OD/OC and lesion segmentation tasks. The learned policies are empirically validated to be model-agnostic and can transfer well to other models. The source code is available at https://github.com/CRazorback/AADG.