29 May 2024 in IEEE Journal of Biomedical and Health Informatics
The Segment Anything Model (SAM) is a foundational model that has demonstrated impressive results in the field of natural image segmentation. However, its performance remains suboptimal for medical image segmentation, particularly when delineating lesions with irregular shapes and low contrast. This can be attributed to the significant domain gap between medical images and natural images on which SAM was originally trained. In this paper, we propose an adaptation of SAM specifically tailored for lesion segmentation termed LeSAM. LeSAM first learns medical-specific domain knowledge through an efficient adaptation module and integrates it with the general knowledge obtained from the pre-trained SAM. Subsequently, we leverage this merged knowledge to generate lesion masks using a modified mask decoder implemented as a lightweight U-shaped network design. This modification enables better delineation of lesion boundaries while facilitating ease of training. We conduct comprehensive experiments on various lesion segmentation tasks involving different image modalities such as CT scans, MRI scans, ultrasound images, dermoscopic images, and endoscopic images. Our proposed method achieves superior performance compared to previous state-of-the-art methods in 8 out of 12 lesion segmentation tasks while achieving competitive performance in the remaining 4 datasets. Additionally, ablation studies are conducted to validate the effectiveness of our proposed adaptation modules and modified decoder.
26 May 2023 in IEEE Transactions on Medical Imaging
Low-dose computed tomography (LDCT) imaging faces great challenges. Although supervised learning has revealed great potential, it requires sufficient and high-quality references for network training. Therefore, existing deep learning methods have been sparingly applied in clinical practice. To this end, this paper presents a novel Unsharp Structure Guided Filtering (USGF) method, which can reconstruct high-quality CT images directly from low-dose projections without clean references. Specifically, we first employ low-pass filters to estimate the structure priors from the input LDCT images. Then, inspired by classical structure transfer techniques, deep convolutional networks are adopted to implement our imaging method which combines guided filtering and structure transfer. Finally, the structure priors serve as the guidance images to alleviate over-smoothing, as they can transfer specific structural characteristics to the generated images. Furthermore, we incorporate traditional FBP algorithms into self-supervised training to enable the transformation of projection domain data to the image domain. Extensive comparisons and analyses on three datasets demonstrate that the proposed USGF has achieved superior performance in terms of noise suppression and edge preservation, and could have a significant impact on LDCT imaging in the future.
8 September 2025 in Engineering Applications of Artificial Intelligence
3 April 2026 in Biomedical Signal Processing and Control
4 May 2026 in SSRN Electronic Journal
Click-based interactive segmentation aims to delineate a target object with minimal user effort by progressively refining the mask from sparse clicks. Although many recent methods feed the previous prediction (probability map) back into the network, the probability representation is often treated as a weak auxiliary cue, and semantic consistency with the evolving feature representation is rarely enforced. This mismatch tends to accumulate across iterations, introducing redundant signals and weakening refinement in complex scenes, which ultimately demands extra corrective clicks. To address this, we propose DRSM, a Dual-Representation Collaborative Modulation (DRCM) that promotes deep cross-iteration collaboration between probability and feature maps via region-focused refinement and explicit representation alignment. Specifically, DRSM performs bidirectional semantic alignment to stabilize cross-iteration fusion, while a density-driven Region-Focused Strategy (RFS) emphasizes user-indicated areas and suppresses irrelevant interference. Building on this guidance, a Dual-Modulation Enhancement (DME) scheme jointly calibrates the previous probability map and the current multi-scale features, yielding more reliable refinement under sparse interactions. Extensive experiments on five public benchmarks and three out-of-domain datasets show that DRSM achieves accurate and reliable segmentation with fewer clicks, while generalizing robustly to unseen domains.