Salvato in:
Dettagli Bibliografici
Autori principali: Bhuiyan, Mohammad Robaitul Islam, Bhat, Sheethal, Qahqaie, Melika, Nguyen, Tri-Thien, Perez-Toro, Paula Andrea, Arias-Vergara, Tomas, Maier, Andreas
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2603.17576
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915894633955328
author Bhuiyan, Mohammad Robaitul Islam
Bhat, Sheethal
Qahqaie, Melika
Nguyen, Tri-Thien
Perez-Toro, Paula Andrea
Arias-Vergara, Tomas
Maier, Andreas
author_facet Bhuiyan, Mohammad Robaitul Islam
Bhat, Sheethal
Qahqaie, Melika
Nguyen, Tri-Thien
Perez-Toro, Paula Andrea
Arias-Vergara, Tomas
Maier, Andreas
contents Precise localization and delineation of brain tumors using Magnetic Resonance Imaging (MRI) are essential for planning therapy and guiding surgical decisions. However, most existing approaches rely on task-specific supervised models and are constrained by the limited availability of annotated data. To address this, we propose LoGSAM, a parameter-efficient, detection-driven framework that transforms radiologist dictation into text prompts for foundation-model-based localization and segmentation. Radiologist speech is first transcribed and translated using a pretrained Whisper ASR model, followed by negation-aware clinical NLP to extract tumor-specific textual prompts. These prompts guide text-conditioned tumor localization via a LoRA-adapted vision-language detection model, Grounding DINO (GDINO). The LoRA adaptation updates using 5% of the model parameters, thereby enabling computationally efficient domain adaptation while preserving pretrained cross-modal knowledge. The predicted bounding boxes are used as prompts for MedSAM to generate pixel-level tumor masks without any additional fine-tuning. Conditioning the frozen MedSAM on LoGSAM-derived priors yields a state-of-the-art dice score of 80.32% on BRISC 2025. In addition, we evaluate the full pipeline using German dictations from a board-certified radiologist on 12 unseen MRI scans, achieving 91.7% case-level accuracy. These results highlight the feasibility of constructing a modular, speech-to-segmentation pipeline by intelligently leveraging pretrained foundation models with minimal parameter updates.
format Preprint
id arxiv_https___arxiv_org_abs_2603_17576
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LoGSAM: Parameter-Efficient Cross-Modal Grounding for MRI Segmentation
Bhuiyan, Mohammad Robaitul Islam
Bhat, Sheethal
Qahqaie, Melika
Nguyen, Tri-Thien
Perez-Toro, Paula Andrea
Arias-Vergara, Tomas
Maier, Andreas
Computer Vision and Pattern Recognition
Precise localization and delineation of brain tumors using Magnetic Resonance Imaging (MRI) are essential for planning therapy and guiding surgical decisions. However, most existing approaches rely on task-specific supervised models and are constrained by the limited availability of annotated data. To address this, we propose LoGSAM, a parameter-efficient, detection-driven framework that transforms radiologist dictation into text prompts for foundation-model-based localization and segmentation. Radiologist speech is first transcribed and translated using a pretrained Whisper ASR model, followed by negation-aware clinical NLP to extract tumor-specific textual prompts. These prompts guide text-conditioned tumor localization via a LoRA-adapted vision-language detection model, Grounding DINO (GDINO). The LoRA adaptation updates using 5% of the model parameters, thereby enabling computationally efficient domain adaptation while preserving pretrained cross-modal knowledge. The predicted bounding boxes are used as prompts for MedSAM to generate pixel-level tumor masks without any additional fine-tuning. Conditioning the frozen MedSAM on LoGSAM-derived priors yields a state-of-the-art dice score of 80.32% on BRISC 2025. In addition, we evaluate the full pipeline using German dictations from a board-certified radiologist on 12 unseen MRI scans, achieving 91.7% case-level accuracy. These results highlight the feasibility of constructing a modular, speech-to-segmentation pipeline by intelligently leveraging pretrained foundation models with minimal parameter updates.
title LoGSAM: Parameter-Efficient Cross-Modal Grounding for MRI Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.17576