Anatomy-Guided Foundation Model Adaptation with Within-Case Prototype Supervision for Standard Plane Detection in Fetal Ultrasound Blind Sweeps

• URL: http://arxiv.org/abs/2608.27051v1
• Date: Thu, 27 Aug 2026 12:36:56 GMT
• Title: Anatomy-Guided Foundation Model Adaptation with Within-Case Prototype Supervision for Standard Plane Detection in Fetal Ultrasound Blind Sweeps
• Abstract summary: AnatoProto is a sequence-level framework that adapts a frozen BiomedCLIP encoder to fetal blind sweeps.<n>On the ACOUSLIC-AI benchmark, AnatoProto reaches a test F1 of 67.72, outperforming the strongest foundation-model baseline.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Detecting the fetal abdominal circumference standard plane in low-cost obstetric blind sweeps is a highly imbalanced frame-classification problem: positive frames account for under 3% of a sequence, form short contiguous segments, and are poorly handled by off-the-shelf ultrasound and vision foundation models. We propose AnatoProto, a lightweight sequence-level framework that adapts a frozen BiomedCLIP encoder to fetal blind sweeps through four components: (i) anatomy-weighted spatial pooling that uses nnU-Net abdominal-region probabilities as a spatial prior to reweight BiomedCLIP patch tokens, so frozen semantic features are aggregated onto anatomically meaningful regions; (ii) a within-case prototype loss that pulls each frame embedding toward the mean of positive frames of the same sweep, exploiting case-level structure unavailable at the frame level; (iii) a three-stage cascade refinement (frame->segment->case-level rejecter) that lifts the prediction unit from noisy frames to structurally-constrained segments; and (iv) a hybrid prediction head that jointly models per-frame stability and inter-frame boundary transitions to suppress boundary false positives. On the ACOUSLIC-AI benchmark, AnatoProto reaches a test F1 of 67.72, outperforming the strongest foundation-model baseline (FetalCLIP + PRS, F1 = 54.52) by +13.20 F1 and the strongest video temporal-action-detection baseline (TriDet + PRS) by +15.76 F1. A synergy study, backed by embedding geometry and paired-bootstrap confidence intervals, shows that the prototype loss and anatomy-weighted pooling are not additive: applied alone the prototype loss reduces recall by 12 points, but combined with anatomy-weighted pooling it increases recall by 6.5 points -- a sign-flip we trace to the accuracy of the within-case prototype.

Related papers

• Bridging Vision Foundation Model Priors with CLIP for Spatial-aware Few-shot Anomaly Detection in Medical Images [48.646448077284276]
  We propose Spatial-FAD, a framework for effective few-shot medical anomaly detection.<n>It improves lesion localization by combining VFM spatial priors with CLIP semantics.<n>Our method achieves an average improvement of over 11.4% in Dice score and 1.8% in AUC.
  arXiv  Detail & Related papers  (2026-09-11T05:21:01Z)
• Test-time Adaptation of Pelvic Bone Segmentation Models via Dynamic Reliability-Guided [62.10276009853225]
  Reliable pelvic bone segmentation from CT is essential for robot-assisted pelvic trauma surgery.<n>Existing methods show limited effectiveness for PBS.<n>We propose a novel closed-loop dynamic Reliability-Guided framework (ReGA) for PBS.
  arXiv  Detail & Related papers  (2026-08-01T08:13:20Z)
• AuricularWorld: Hierarchical Action-Guided World Modeling for Fine-Grained Auricular Structure Segmentation from CT Scans [4.139833999494916]
  We propose a world-model-based segmentation framework that enables iterative anatomical reasoning.<n>We show that the proposed framework consistently improves segmentation accuracy and reduces HD95 by more than 43% for small, irregular, and overlapping auricular structures in CT.
  arXiv  Detail & Related papers  (2026-07-30T16:38:30Z)
• Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems [47.027290803102666]
  Normal behaviour is the union of many imbalanced, curved, thin-fringed operating regimes.<n>We model the normal law with a jointly learned latent representation plus explicit Gaussian-mixture mode clustering.<n>We evaluate under a deliberately fair protocol: raw point-wise metrics with no point adjustment, a trivial-detector difficulty split, prevalence-matched F1, and train-normal-only calibration.
  arXiv  Detail & Related papers  (2026-07-07T10:10:00Z)
• Validation of Whole-Slide Foundation Models for Image Retrieval in TCGA Data [1.834937230572996]
  We benchmarked ten pipelines on 9,387 diagnostic slides spanning 17 organs and 60 diagnoses from The Cancer Genome Atlas.<n>Methods included four pre-trained slide foundation models, a supervised attention-based multiple instance learning (ABMIL) aggregator on patch embeddings, and patch-level retrieval.
  arXiv  Detail & Related papers  (2026-04-28T19:19:53Z)
• CrossPan: A Comprehensive Benchmark for Cross-Sequence Pancreas MRI Segmentation and Generalization [15.982441239969354]
  CrossPan is a benchmark comprising 1,386 3D scans across three routinely acquired sequences.<n>Models achieving Dice scores above 0.85 in-domain collapse to near-zero when transferred across sequences.<n>Semi-supervised learning offers gains only under stable intensity distributions and becomes unstable on generalization sequences with high intra-organ variability.
  arXiv  Detail & Related papers  (2026-04-20T20:00:32Z)
• Architecture-Agnostic Modality-Isolated Gated Fusion for Robust Multi-Modal Prostate MRI Segmentation [0.0]
  Multi-weighted prostate MRI combines T2-parametric (T2W), apparent diffusion coefficient (ADC), and high b-value diffusion-weighted (HBV) sequences.<n>In practice, the diffusion sequences are more frequently subject to acquisition variability, motion, and artifacts than T2W.<n>We propose Modality-Isolated Gated Fusion (MIGF) to maintain separate modality-specific encoding streams before a learned gating stage.
  arXiv  Detail & Related papers  (2026-04-12T15:54:21Z)
• PRIME: Prototype-Driven Multimodal Pretraining for Cancer Prognosis with Missing Modalities [86.63247982275396]
  PRIME is a missing-aware multimodal self-supervised pretraining framework.<n>It learns robust and transferable representations from partially observed cohorts.<n>We evaluate PRIME on The Cancer Genome Atlas with label-free pretraining on 32 cancer types.
  arXiv  Detail & Related papers  (2026-04-05T21:14:27Z)
• Multi-head automated segmentation by incorporating detection head into the contextual layer neural network [0.0]
  We propose a gated multi-head Transformer architecture based on Swin U-Net, augmented with inter-slice context integration.<n>We show that the gated model substantially outperforms a non-gated segmentation-only baseline.<n>These results indicate that detection-based gating enhances robustness and anatomical plausibility in automated segmentation applications.
  arXiv  Detail & Related papers  (2026-02-02T18:51:25Z)
• A Dual-Mode ViT-Conditioned Diffusion Framework with an Adaptive Conditioning Bridge for Breast Cancer Segmentation [2.8110777003480654]
  In breast ultrasound images, precise lesion segmentation is essential for early diagnosis.<n>Standard convolutional architectures frequently fall short in capturing enough global context.<n>We suggest a flexible, conditional Denoising Diffusion Model that combines an enhanced UNet-based generative decoder with a Vision Transformer (ViT) encoder.
  arXiv  Detail & Related papers  (2025-11-08T12:33:18Z)
• VesSAM: Efficient Multi-Prompting for Segmenting Complex Vessel [68.24765319399286]
  We present VesSAM, a powerful and efficient framework tailored for 2D vessel segmentation.<n>VesSAM integrates (1) a convolutional adapter to enhance local texture features, (2) a multi-prompt encoder that fuses anatomical prompts, and (3) a lightweight mask decoder to reduce jagged artifacts.<n>VesSAM consistently outperforms state-of-the-art PEFT-based SAM variants by over 10% Dice and 13% IoU.
  arXiv  Detail & Related papers  (2025-11-02T15:47:05Z)
• 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)
• Fast and Accurate Blind Flexible Docking [79.88520988144442]
  Molecular docking that predicts the bound structures of small molecules (ligands) to their protein targets plays a vital role in drug discovery.<n>We propose FABFlex, a fast and accurate regression-based multi-task learning model designed for realistic blind flexible docking scenarios.
  arXiv  Detail & Related papers  (2025-02-20T07:31:13Z)
• 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.