Semi-supervised learning (SSL) has achieved notable progress in medical image segmentation. To achieve effective SSL, a model needs to be able to efficiently learn from limited labeled data and effectively exploit knowledge from abundant unlabeled data. Recent developments in visual foundation models, such as the Segment Anything Model (SAM), have demonstrated remarkable adaptability with improved sample efficiency. To seamlessly harness foundation models in SSL, we propose a SAM-driven cross prompting framework with adaptive sampling and prompt consistency for semi-supervised medical image segmentation, named CPAC-SAM. Our method employs SAM's unique prompt design and innovates a cross prompting strategy within a dual-branch framework to automatically generate prompts and supervision across two decoder branches, enabling effective learning from both scarce labeled and valuable unlabeled data. To ensure the quality of prompts for unlabeled data and provide meaningful supervision in the cross prompting scheme, we propose an innovative prototype-guided grid sampling strategy with adaptive intervals to simultaneously improve the reliability of the prompt selection area and ensure both adequate prompt density and complete target coverage. We further design a novel prompt consistency regularization to reduce SAM's prompt sensitivity and to enhance the output invariance under different prompts. We validate our method on five medical image segmentation tasks, encompassing both 2D and 3D scenarios. The extensive experiments with different labeled-data ratios and modalities demonstrate the superiority of our proposed method over the state-of-the-art SSL methods, with more than 4.1% and 3.8% Dice improvement on the breast cancer segmentation task and left atrium segmentation task, respectively. Our code is available at: https://github.com/JuzhengMiao/CPAC-SAM.

Deep feature-based approaches for brain tumor classification and segmentation in medical imaging

In radiotherapy, medical image segmentation is performed to achieve a more structured view of the patient’s anatomical region. Automatic segmentation methods aim to eliminate the significant time investment required by the manual and semi-automatic segmentation processes that are most commonly used today. The present work describes an innovative seed-based automatic segmentation method for computed tomography (CT) images, known as LUNg Automatic Seeding and Segmentation (LUNAS). The study compares LUNAS with other segmentation algorithms from the Lung CT Segmentation Challenge (LCTSC), which are based on neural networks and multi-atlas approaches. The findings indicate that LUNAS achieved a Dice accuracy metric of 0.96 and 0.97 for the left and right lungs, respectively, matched the top-performing DL methods in the competition. Additionally, other state-of-the-art methods were evaluated for comparison, including one seed-based method similar to LUNAS, as well as a deep learning method. Using three other public thoracic CT image datasets, a detailed and fair analysis was performed to compare the algorithms indicating the effectiveness of LUNAS. LUNAS also has the ability to segment other regions, such as the trachea, bones, and skin. • LUNAS achieves state-of-the-art accuracy in lung CT segmentation, outper-forming traditional methods and matching deep learning approaches. • The method employs a novel automatic seed generation strategy combined with the Relaxed Oriented Image Foresting Transform (ROIFT). • LUNAS provides robust segmentation without requiring GPU acceleration, making it computationally efficient. • The method demonstrates superior adaptability, performing well on multiple publicly available thoracic CT datasets. • The methodology is adaptable to other anatomical structures such as trachea, bones, and skin.

Shape-aware medical image segmentation via frequency domain partitioning

Enhancing feature fusion of U-like networks with dynamic skip connections.

SAM2-driven dual-teacher framework using hierarchical cross-slice context for semi-supervised 3D medical image segmentation

KP2L: Knowledge-driven pyramid prototype learning for semi-supervised medical image segmentation

Dynamic ambiguity perception for semi-supervised medical image segmentation

DFEN: Dual feature equalization network for medical image segmentation

Efficient 3D medical image segmentation via Single-Slice Scribble Annotations: A near unsupervised approach

EGNet: A boundary-region closed-loop network for medical image segmentation with fuzzy lesions

Aggregate twice more efficiently: Dual feature aggregation transformer for medical image segmentation

Window-xLSTMUnet: A medical image segmentation network fusing improved xLSTM with multi-scale receptive fields

FedCA: Federated domain generalization for medical image segmentation via cross-client feature style transfer and adaptive style alignment

Heterogeneous multi-network cross pseudo-supervised medical image segmentation

Domains shift originated from differences in devices or patients in the medical field, poses a significant challenge when applying pre-trained models to clinical applications. To tackle this challenge, domain adaptation methods have been explored. However, most existing methods are designed for a single target domain adaptation or require sharing all target domain data for adaptation, which is infeasible in the medical field due to privacy issues. In this paper, we propose a novel unsupervised multi-target domain adaptation method without requiring data sharing. To this end, we introduce an additional signal, termed Adaptogen-Perturbation (AP) optimized to bridge the gap between the source and target domains. The optimized AP is injected into the latent feature and facilitates the adaptation of the pre-trained model to the target domain. Moreover, we propose a Spectral/Geometric Consistency learning framework to optimize the AP in an unsupervised manner. This promotes consistent predictions across two types of transformations: geometric and frequency-space spectral transformations, enhancing robustness to both variations. Extensive experiments with multiple medical segmentation datasets demonstrate the effectiveness of APs.

Sparse prior guided decoder network for medical image segmentation

Enhancing medical image segmentation with collaborative and contrastive learning in mixed-domain settings

DyGLNet: Hybrid global-local feature fusion with dynamic upsampling for medical image segmentation

Progressive cross-scale semantic alignment for language-guided medical image segmentation