arXiv Computer Vision

BrainIAC: Interactive 3D Brain Lesion Segmentation across Heterogeneous MRI Modalities with Online Adaptation

BrainIAC is a unified framework for 3D brain lesion segmentation that handles heterogeneous MRI modalities and adapts online during interactive segmentation. It combines a multi‑modal backbone trained with zero‑filling and random modality dropping, 3D interactive prompts that default to fully automatic predictions, and a two‑stage online adaptation guided by pseudo‑labels and a Click‑Centered Gaussian loss. Experiments on seven MRI datasets show that the components work synergistically, outperforming existing methods and generalizing to unseen modalities and pathologies.

arXiv Computer Vision
Aug 27

Synergistic Modality-and-Slice Memory Framework for Cross-Modal 3D Brain Tumor Segmentation

The paper introduces MSM‑Seg, a dual‑memory segmentation framework for 3D multi‑modal brain tumor segmentation. It combines a modality‑and‑slice memory attention module to capture cross‑modal and spatial‑slice dependencies, a multi‑scale category‑agnostic prompt encoder for whole‑tumor guidance, and a modality‑adaptive fusion decoder to integrate complementary decoding information. Experiments on various MRI datasets show that MSM‑Seg surpasses state‑of‑the‑art methods for metastases and glioma tumor segmentation.

By Yuxiang Luo, Qing Xu, Hai Huang, Yuqi Ouyang, Xiangjian He, Zhen Chen, Wenting Duan, Jiebo Luo
arXiv Machine Learning
Jun 16

Lesion-DDPM: Lesion-Enhanced 3D Diffusion for MS MRI Synthesis

arXiv:2606. 15457v1 Announce Type: cross Abstract: 3D FLAIR MRI is widely recommended as one of the standard MRI sequences for brain imaging in multiple sclerosis (MS), but publicly available MS datasets remain relatively small and vary across scanners, acquisition protocols, and lesion patterns.

By Weidong Zhang, Yongchan Jung, Shafayat Mowla Anik, Furen Xiao, Vasudevan Janarthanan, Enkhzaya Chuluunbaatar, Byeong Kil Lee, Jeeho Ryoo
arXiv Machine Learning
Oct 21

Convergent Performance, Divergent Mechanisms in LoRA-Adapted Vision Foundation Models for Brain MRI

The study examines how LoRA adaptation of four Vision Foundation Models—DINOv3, SigLIP2, MerMED-FM, and BiomedCLIP—affects brain MRI tasks. Despite achieving similar segmentation accuracy (mean Dice 78‑81%), the models exhibit distinct internal dynamics, with early‑to‑mid layers being most crucial for segmentation. Restricting LoRA to these layers yields comparable performance while mitigating negative impacts on Alzheimer’s disease classification.

By Cecilia Diana-Albelda, Fiona Kekwick, Raghav Mehta, Aleksandra Kulbaka, Paul M. Matthews, Jesus Bescos, \'Alvaro Garc\'ia-Mart\'in, Wenjia Bai, Bernhard Kainz
arXiv Computer Vision
Oct 21

CAPS: Clinically-Verifiable Anatomy-Prioritized Synthesis for Missing MRI Imputation in Brain Tumor Segmentation

CAPS is a Clinically‑Verifiable Anatomy‑Prioritized Synthesis framework designed to impute missing MRI modalities for brain tumor segmentation. It uses a hybrid loss (L₁ + L_AFP + L_DualDice) to preserve deep anatomical features and topological boundaries, achieving high synthesis fidelity and robust segmentation (82.38% mean Dice) on the BraTS Adult Glioma dataset. The method also generalizes well to unseen cohorts and has been validated by blinded radiologists, who report improved image quality and clinical utility.

By Cecilia Diana-Albelda, Arthur Longuefosse, Mario Rub\'en Paz Campos, \'Alvaro Garc\'ia-Mart\'in, Jesus Bescos
arXiv Computer Vision
6d ago

CRIL-U-Net: Compact Ratio-Interaction Learning for Focal Cortical Dysplasia Segmentation from T1w and FLAIR MRI

CRIL-U-Net is a 3D U‑Net variant that introduces a Compact Ratio‑Interaction Learning module to fuse T1‑weighted and FLAIR MRI modalities for focal cortical dysplasia (FCD) type II segmentation. In a five‑fold cross‑validation on 85 FCD subjects and 25 controls, CRIL‑U‑Net achieved a mean Dice score of 0.196±0.262 using a focal Tversky‑Focal loss, outperforming a baseline 3D U‑Net and an attention‑augmented U‑Net while adding only 4,320 parameters. The model localized 44 of 85 lesions, a significant improvement over the other architectures, though a 48.2 % miss rate remains.

By Soumen Ghosh, Subhojit Mandal, Tilottama Goswami, Amit Soni Arya, John Phamnguyen, Rajat Vashistha
arXiv AI
Jun 12

BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning

arXiv:2604. 27277v3 Announce Type: replace-cross Abstract: Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data.

By Yizhou Wu, Shansong Wang, Yuheng Li, Mojtaba Safari, Mingzhe Hu, Chih-Wei Chang, Harini Veeraraghavan, Xiaofeng Yang
arXiv AI
Aug 21

MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal Prostate MRI Segmentation

arXiv:2510. 17529v3 Announce Type: replace-cross Abstract: Active Surveillance (AS) is a treatment option for managing low and intermediate-risk prostate cancer (PCa), aiming to avoid overtreatment while monitoring disease progression through serial MRI and clinical follow-up.

By Yovin Yahathugoda, Davide Prezzi, Patricia A. Gutierrez, Piyalitt Ittichaiwong, Vicky Goh, Sebastien Ourselin, Michela Antonelli