Parameter-Efficient Adaptation of SAM 3 for Automated ITV Generation from 4DCT Images

• URL: http://arxiv.org/abs/2606.15604v1
• Date: Sun, 14 Jun 2026 05:12:57 GMT
• Title: Parameter-Efficient Adaptation of SAM 3 for Automated ITV Generation from 4DCT Images
• Abstract summary: Four-dimensional computed tomography (4DCT) captures the full respiratory cycle of thoracic anatomy.<n>Current Internal Target Volume contouring process each phase in isolation, discarding temporal coherence and leaving vulnerable to phase-specific artifacts.<n>We present a framework that applies parameter-efficient fine-tuning to the Segment Anything Model 3 (SAM 3) via low-rank adaptation (LoRA)<n>With only seven annotated volumes, the framework retains over 95% of full-data accuracy, and the entire pipeline is trainable on a single consumer-grade GPU.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Four-dimensional computed tomography (4DCT) captures the full respiratory cycle of thoracic anatomy, yet current Internal Target Volume contouring workflows process each phase in isolation, discarding temporal coherence and leaving contours vulnerable to phase-specific artifacts. We present a lightweight framework that applies parameter-efficient fine-tuning to the Segment Anything Model 3 (SAM 3) via low-rank adaptation (LoRA) to align its text-prompted segmentation with the medical domain using only seven annotated 3D CT volumes. Furthermore, the framework incorporates a hard negative mining strategy to improve boundary discrimination in low-contrast thoracic regions. At inference, phase-wise predictions are refined through phase-coherent temporal filtering and spatial connectivity analysis. Since respiratory motion is continuous and periodic, genuine anatomy appears in contiguous blocks of phases, whereas transient artifacts appear sporadically and are thus effectively suppressed. Experiments on pulmonary and cardiac structures yield median Dice scores of 0.968 and 0.910 with 95th-percentile Hausdorff distances of 0.998 mm and 2.931 mm, respectively. The proposed framework effectively eliminates the severe false-positive predictions inherent in the zero-shot inference of the unadapted SAM 3. With only seven annotated volumes, the framework retains over 95% of full-data accuracy, and the entire pipeline is trainable on a single consumer-grade GPU, demonstrating a scalable, data-efficient solution for adaptive radiotherapy.

Related papers

• AsymFeX: A Symmetry-Driven Framework for Ischemic Stroke Segmentation Across Imaging Modalities and Stroke Stages [0.8886706641070187]
  Acute Ischemic Stroke (AIS) lesions are essential for stroke prognosis and treatment planning.<n>Non-contrast CT (NCCT) exhibits subtle infarct contrast, making manual delineation slow and labor-intensive.<n>We propose a two-stage, nnU-Net-compatible 3D segmentation method.
  arXiv  Detail & Related papers  (2026-08-20T08:12:31Z)
• EPC-3D-Diff: Equivariant Physics Consistent Conditional 3D Latent Diffusion for CBCT to CT Synthesis [36.790666885042356]
  We propose EPC-3D-Diff, a novel conditional 3D latent diffusion framework for CBCT to CT synthesis.<n>We exploit the fact that an in plane rotation of the volume corresponds to an angular shift in its projections.<n>We validate on a paired head CBCT/CT phantom dataset, including repeat scans, and paired clinical data using patient wise splits.<n> EPC-3D-Diff generalizes well and achieved substantial improvements, +7.4 dB (phantom) and +1.8 dB (clinical data) in PSNR compared to state of the art methods.
  arXiv  Detail & Related papers  (2026-05-19T20:34:09Z)
• Virtual Intraoperative CT (viCT): Sequential Anatomic Updates for Modeling Tissue Resection Throughout Endoscopic Sinus Surgery [0.8060701341619877]
  We present Virtual Intraoperative CT (viCT), a method for sequentially updating preoperative CT throughout endoscopic sinus surgery (ESS)<n>viCT uses 3D reconstructions from monocular endoscopic video to enable visualization of evolving anatomy in CT format.<n>Results: viCT updates show agreement with ground-truth anatomy across surgical stages, with submillimeter mean surface errors.
  arXiv  Detail & Related papers  (2026-03-07T00:23:05Z)
• DRBD-Mamba for Robust and Efficient Brain Tumor Segmentation with Analytical Insights [54.87947751720332]
  Accurate brain tumor segmentation is significant for clinical diagnosis and treatment.<n>Mamba-based State Space Models have demonstrated promising performance.<n>We propose a dual-resolution bi-directional Mamba that captures multi-scale long-range dependencies with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-10-16T07:31:21Z)
• Depth-Sequence Transformer (DST) for Segment-Specific ICA Calcification Mapping on Non-Contrast CT [38.85617601239779]
  Conventional 3D models are forced to process downsampled volumes or isolated patches.<n>We reformulate the 3D challenge as a textbfParallel Probabilistic Landmark localization task along the 1D axial dimension.<n>We propose the textbfDepth-Sequence Transformer (DST), a framework that processes full-resolution CT volumes as sequences of 2D slices.
  arXiv  Detail & Related papers  (2025-07-10T23:12:12Z)
• TRACE: Temporally Reliable Anatomically-Conditioned 3D CT Generation with Enhanced Efficiency [40.82927972746919]
  TRACE is a framework that generates 3D medical images with temporal alignment.<n>An overlapping-frame frame pairs pairs into a flexible length sequence, reconstructed into atemporally and anatomically aligned 3D volume.
  arXiv  Detail & Related papers  (2025-07-01T14:35:39Z)
• Lung-DETR: Deformable Detection Transformer for Sparse Lung Nodule Anomaly Detection [0.0]
  Accurate lung nodule detection for computed tomography (CT) scan imagery is challenging in real-world settings. We introduce a novel solution leveraging custom data preprocessing and Deformable Detection Transformer (Deformable- DETR) A 7.5mm Maximum Intensity Projection (MIP) is utilized to combine adjacent lung slices into single images, reducing the slice count and decreasing sparsity. Our model achieves state-of-the-art performance on the LUNA16 dataset with an F1 score of 94.2% (95.2% recall, 93.3% precision)
  arXiv  Detail & Related papers  (2024-09-08T19:24:38Z)
• Weakly supervised segmentation of intracranial aneurysms using a novel 3D focal modulation UNet [0.5106162890866905]
  We propose FocalSegNet, a novel 3D focal modulation UNet, to detect an aneurysm and offer an initial, coarse segmentation of it from time-of-flight MRA image patches. We trained and evaluated our model on a public dataset, and in terms of UIA detection, our model showed a low false-positive rate of 0.21 and a high sensitivity of 0.80.
  arXiv  Detail & Related papers  (2023-08-06T03:28:08Z)
• 3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentation [52.699139151447945]
  We propose a novel adaptation method for transferring the segment anything model (SAM) from 2D to 3D for promptable medical image segmentation. Our model can outperform domain state-of-the-art medical image segmentation models on 3 out of 4 tasks, specifically by 8.25%, 29.87%, and 10.11% for kidney tumor, pancreas tumor, colon cancer segmentation, and achieve similar performance for liver tumor segmentation.
  arXiv  Detail & Related papers  (2023-06-23T12:09:52Z)
• Validated respiratory drug deposition predictions from 2D and 3D medical images with statistical shape models and convolutional neural networks [47.187609203210705]
  We aim to develop and validate an automated computational framework for patient-specific deposition modelling. An image processing approach is proposed that could produce 3D patient respiratory geometries from 2D chest X-rays and 3D CT images.
  arXiv  Detail & Related papers  (2023-03-02T07:47:07Z)
• An Efficient and Robust Method for Chest X-Ray Rib Suppression that Improves Pulmonary Abnormality Diagnosis [0.49998148477760956]
  Suppression of thoracic bone shadows on chest X-rays (CXRs) has been indicated to improve the diagnosis of pulmonary disease. Previous approaches can be categorized as unsupervised physical and supervised deep learning models. We propose a generalizable yet efficient workflow of two stages: (1) training pairs generation with GT bone shadows eliminated in minimization by a physical model in spatially transformed gradient fields. (2) fully supervised image denoising network training on stage-one datasets for fast rib removal on incoming CXRs.
  arXiv  Detail & Related papers  (2023-02-19T23:47:02Z)
• Pulmonary Vessel Segmentation based on Orthogonal Fused U-Net++ of Chest CT Images [1.8692254863855962]
  We present an effective framework and refinement process of pulmonary vessel segmentation from chest computed tomographic (CT) images. The key to our approach is a 2.5D segmentation network applied from three axes, which presents a robust and fully automated pulmonary vessel segmentation result. Our method outperforms other network structures by a large margin and achieves by far the highest average DICE score of 0.9272 and precision of 0.9310.
  arXiv  Detail & Related papers  (2021-07-03T21:46:29Z)
• SCPM-Net: An Anchor-free 3D Lung Nodule Detection Network using Sphere Representation and Center Points Matching [47.79483848496141]
  We propose a 3D sphere representation-based center-points matching detection network (SCPM-Net) It is anchor-free and automatically predicts the position, radius, and offset of nodules without the manual design of nodule/anchor parameters. We show that our proposed SCPM-Net framework achieves superior performance compared with existing used anchor-based and anchor-free methods for lung nodule detection.
  arXiv  Detail & Related papers  (2021-04-12T05:51:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.