Hi-Seg achieves a mean Dice score of nearly 85% for pulmonary nodule segmentation by having humans iteratively refine prompts for the Segment Anything Model, outperforming standalone deep learning and SAM models on a large multi-center dataset.
hub
arXiv preprint arXiv:2408.00874 (2024)
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
SAM-family models split into occluder-aware types that avoid predicting into occluded regions and occluder-agnostic types that confidently segment hidden areas, shown via a new benchmark on polyp datasets.
BadVSFM is the first effective backdoor attack on prompt-driven video segmentation foundation models, using a two-stage encoder-decoder strategy to achieve high attack success rates with limited clean performance loss.
Consispace is a semantic-aware resampling method that uses an implicit neural network with ODE constraints and feature reweighting to achieve consistent axial voxel spacing while preserving anatomy and semantics, improving downstream segmentation.
Tractography from ex vivo dMRI is used as a generative prior to create synthetic patches that, when mixed with real data and domain randomization, train a U-Net to segment fiber bundles in tracer histology with 3x less manual annotation.
A multimodal training pipeline with phonological bounding-box priors and cross-modal contrastive alignment transfers speech supervision to single-modality rtMRI vocal tract segmentation and outperforms prior methods on two datasets.
A four-phase multi-agent co-scientist tests natural-language hypotheses on cardiac and glioma MRI and labels outcomes Supported, Refuted, Underpowered, or Invalid with an executable evidence trail.
A training-free retrieval-to-prompt method using DINOv3 correspondences and SAM2 reports 0.9863 mIoU on mixed tongue images.
GleSAM integrates latent diffusion into SAM and SAM2 to boost segmentation robustness on low-quality images using minimal extra parameters and a new LQSeg dataset.
A synthetic-data-driven, hierarchy-aware adaptation of foundation models produces geometry-consistent representations that improve pose estimation and monocular depth in endoscopy.
A lightweight MLP reads a 5x5 patch of MedSAM's image embedding around one click, predicts a bounding box, and uses it as a spatial prompt, improving Dice by about 0.5-1.5 points on CT/MRI/ultrasound benchmarks.
CAR-SAM introduces MatMul-Aware Compensation and Joint Cross-Attention Reconstruction to enable stable 4-bit post-training quantization of SAM, outperforming prior PTQ methods by 14.6% mAP on SAM-B and 6.6% on SAM-L.
A scale-robust lightweight CNN for glottis segmentation achieves 92.9% mDice at over 170 FPS with a 19 MB model size on three datasets.
SGP-SAM transfers 3D SAM to lesion segmentation using a self-gated module for conditional multi-scale enhancement and a Zoom Loss, achieving 7.3% mDice gain over fine-tuning on MSD Liver Tumor data.
GleSAM++ improves SAM robustness on degraded images by using generative enhancement, feature alignment, and adaptive degradation prediction while adding few parameters.
A novel weakly supervised anomaly detection method for brain MRI that uses discriminative dual prompt tuning for pseudo masks and region-aware spatial attention with location-based random embeddings to achieve SOTA results with under 8 million parameters on BraTS and MSD datasets.
LoRA-adapted SAM 3 with hard-negative mining and phase-coherent filtering achieves median Dice 0.968 on pulmonary structures from 4DCT using seven annotated volumes.
A fine-tuned 3D foundation segmentation model combined with cross pseudo supervision achieves robust liver segmentation across labeled and unlabeled multi-phase, multi-vendor MRI without spatial registration.
citing papers explorer
-
Human and AI collaboration for pulmonary nodule segmentation
Hi-Seg achieves a mean Dice score of nearly 85% for pulmonary nodule segmentation by having humans iteratively refine prompts for the Segment Anything Model, outperforming standalone deep learning and SAM models on a large multi-center dataset.
-
Seeing Through the Tool: A Controlled Benchmark for Occlusion Robustness in Foundation Segmentation Models
SAM-family models split into occluder-aware types that avoid predicting into occluded regions and occluder-agnostic types that confidently segment hidden areas, shown via a new benchmark on polyp datasets.
-
Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models
BadVSFM is the first effective backdoor attack on prompt-driven video segmentation foundation models, using a two-stage encoder-decoder strategy to achieve high attack success rates with limited clean performance loss.
-
Towards Voxel Spacing Consistency for Medical Image Segmentation
Consispace is a semantic-aware resampling method that uses an implicit neural network with ODE constraints and feature reweighting to achieve consistent axial voxel spacing while preserving anatomy and semantics, improving downstream segmentation.
-
Tractography-Driven Synthetic Data Generation for Fiber Bundle Segmentation in Tracer Histology
Tractography from ex vivo dMRI is used as a generative prior to create synthetic patches that, when mixed with real data and domain randomization, train a U-Net to segment fiber bundles in tracer histology with 3x less manual annotation.
-
Speech-Guided Multimodal Learning for Vocal Tract Segmentation in Real-Time MRI
A multimodal training pipeline with phonological bounding-box priors and cross-modal contrastive alignment transfers speech supervision to single-modality rtMRI vocal tract segmentation and outperforms prior methods on two datasets.
-
VERITAS: A Multi-Agent Co-Scientist for Verifiable Image-Derived Hypothesis Testing
A four-phase multi-agent co-scientist tests natural-language hypotheses on cardiac and glioma MRI and labels outcomes Supported, Refuted, Underpowered, or Invalid with an executable evidence trail.
-
Memory-SAM: Human-Prompt-Free Tongue Segmentation via Retrieval-to-Prompt
A training-free retrieval-to-prompt method using DINOv3 correspondences and SAM2 reports 0.9863 mIoU on mixed tongue images.
-
Segment Any-Quality Images with Generative Latent Space Enhancement
GleSAM integrates latent diffusion into SAM and SAM2 to boost segmentation robustness on low-quality images using minimal extra parameters and a new LQSeg dataset.
-
Geometry-Consistent Endoscopic Representations for Image-Guided Navigation via Structured Foundation Model Adaptation
A synthetic-data-driven, hierarchy-aware adaptation of foundation models produces geometry-consistent representations that improve pose estimation and monocular depth in endoscopy.
-
Enhancing MedSAM with a Lightweight Box Predictor for Medical Image Segmentation
A lightweight MLP reads a 5x5 patch of MedSAM's image embedding around one click, predicts a bounding box, and uses it as a spatial prompt, improving Dice by about 0.5-1.5 points on CT/MRI/ultrasound benchmarks.
-
CAR-SAM: Cross-Attention Reconstruction for Post-Training Quantization of the Segment Anything Model
CAR-SAM introduces MatMul-Aware Compensation and Joint Cross-Attention Reconstruction to enable stable 4-bit post-training quantization of SAM, outperforming prior PTQ methods by 14.6% mAP on SAM-B and 6.6% on SAM-L.
-
A Real-time Scale-robust Network for Glottis Segmentation in Nasal Transnasal Intubation
A scale-robust lightweight CNN for glottis segmentation achieves 92.9% mDice at over 170 FPS with a 19 MB model size on three datasets.
-
SGP-SAM: Self-Gated Prompting for Transferring 3D Segment Anything Models to Lesion Segmentation
SGP-SAM transfers 3D SAM to lesion segmentation using a self-gated module for conditional multi-scale enhancement and a Zoom Loss, achieving 7.3% mDice gain over fine-tuning on MSD Liver Tumor data.
-
Towards Any-Quality Image Segmentation via Generative and Adaptive Latent Space Enhancement
GleSAM++ improves SAM robustness on degraded images by using generative enhancement, feature alignment, and adaptive degradation prediction while adding few parameters.
-
RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings for Weakly Supervised Anomaly Detection in Brain MRI Scans
A novel weakly supervised anomaly detection method for brain MRI that uses discriminative dual prompt tuning for pseudo masks and region-aware spatial attention with location-based random embeddings to achieve SOTA results with under 8 million parameters on BraTS and MSD datasets.
-
Parameter-Efficient Adaptation of SAM 3 for Automated ITV Generation from 4DCT Images
LoRA-adapted SAM 3 with hard-negative mining and phase-coherent filtering achieves median Dice 0.968 on pulmonary structures from 4DCT using seven annotated volumes.
-
Label-Efficient Cross-Modality Generalization for Liver Segmentation in Multi-Phase MRI
A fine-tuned 3D foundation segmentation model combined with cross pseudo supervision achieves robust liver segmentation across labeled and unlabeled multi-phase, multi-vendor MRI without spatial registration.
- On Efficient Variants of Segment Anything Model: A Survey