CrossMoDA
CrossMoDA 数据集是一个包含T1加权和T2加权磁共振图像（MRI）及其语义分割标注的医学影像数据集，主要模态为图像数据，用于分割前庭神经鞘瘤和耳蜗这两个关键脑部结构；该数据集旨在支持无监督跨模态域适应技术的研究，特别适用于医学图像分析中的语义分割任务，为前庭神经鞘瘤的随访与治疗规划提供重要的数据基准。
基本信息
资源简介
CrossMoDA
【主题与分类】
主题标签：{''id'': 14708, ''datasetId'': 28052, ''domain'': ''publishDate'', ''task'': ''2021-04-05'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''2021-04-05'', ''level1TagName'': ''2021-04-05'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14707, ''datasetId'': 28052, ''domain'': ''displayName'', ''task'': ''CrossMoDA (Cross-Modality Domain Adaptation)'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''CrossMoDA (Cross-Modality Domain Adaptation)'', ''level1TagName'': ''CrossMoDA (Cross-Modality Domain Adaptation)'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14709, ''datasetId'': 28052, ''domain'': ''publisher'', ''task'': ''crossMoDA-Challenge Team'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''crossMoDA-Challenge Team'', ''level1TagName'': ''crossMoDA-Challenge Team'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14704, ''datasetId'': 28052, ''domain'': ''paperUrl'', ''task'': ''https://arxiv.org/pdf/2201.02831v2.pdf'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''https://arxiv.org/pdf/2201.02831v2.pdf'', ''level1TagName'': ''https://arxiv.org/pdf/2201.02831v2.pdf'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14711, ''datasetId'': 28052, ''domain'': ''publishUrl'', ''task'': ''https://zenodo.org/record/4662239'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''https://zenodo.org/record/4662239'', ''level1TagName'': ''https://zenodo.org/record/4662239'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14705, ''datasetId'': 28052, ''domain'': ''mediaTypes'', ''task'': ''Image'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''Image'', ''level1TagName'': ''Image'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14710, ''datasetId'': 28052, ''domain'': ''taskTypes'', ''task'': ''Semantic Segmentation'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''Semantic Segmentation'', ''level1TagName'': ''Semantic Segmentation'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}、{''id'': 14706, ''datasetId'': 28052, ''domain'': ''labelTypes'', ''task'': ''Semantic_seg_map'', ''filter'': None, ''label'': None, ''dataType'': 0, ''level1Tag'': ''Semantic_seg_map'', ''level1TagName'': ''Semantic_seg_map'', ''level2Tag'': None, ''level2TagName'': None, ''level3Tag'': None, ''level3TagName'': None, ''gmtCreate'': 1720714009, ''gmtModified'': 1720714009, ''tagId'': None}。
【获取与许可】
源站页面：https://modelscope.cn/datasets/OmniData/CrossMoDA
使用许可：[CC BY-SA 4.0]。
下载信息
注册下载
需要注册 Kaggle 账号并登录后下载，适合需要跟踪下载记录和使用 API 的用户。
暂未开放公开下载
无需注册即可直接获取公开样本或文档，适合快速预览和评估数据集质量。
免登录有偿下载
公开数据集受托下载与技术交付服务。
提供高速下载与技术交付服务（收技术服务费，非数据销售）
暂未开放当前数据集主要来源为 Kaggle 公开托管，完整影像包建议通过原始链接或 Kaggle API 获取。
使用方式
数据集获取（ModelScope）
方式一：MsDataset（Python）
# 前置依赖: pip install modelscope
from modelscope.msdatasets import MsDataset
ds = MsDataset.load("OmniData/CrossMoDA", subset_name="default", split="train")
print(ds)
方式二：CLI（命令行）
pip install modelscope
modelscope download dataset OmniData/CrossMoDA
数据集卡片摘录（源站）
displayName: CrossMoDA (Cross-Modality Domain Adaptation)
labelTypes:
-
Semantic_seg_map
license: -
CC BY-SA 4.0
mediaTypes: -
Image
paperUrl: https://arxiv.org/pdf/2201.02831v2.pdf
publishDate: “2021-04-05”
publishUrl: https://zenodo.org/record/4662239
publisher: -
crossMoDA-Challenge Team
tags: -
Vestibular Schwannoma
-
Cochleas
taskTypes: -
Semantic Segmentation
数据集介绍
简介
CrossMoDA 是无监督跨模态 DA 的第一个大型多类基准。挑战的目标是分割与前庭神经鞘瘤 (VS) 的随访和治疗计划相关的两个关键大脑结构：VS 和耳蜗。
引文
@article{dorent2022crossmoda,
title={CrossMoDA 2021 challenge: Benchmark of Cross-Modality Domain Adaptation techniques for Vestibular Schwnannoma and Cochlea Segmentation},
author={Dorent, Reuben and Kujawa, Aaron and Ivory, Marina and Bakas, Spyridon and Rieke, Nicola and Joutard, Samuel and Glocker, Ben and Cardoso, Jorge and Modat, Marc and Batmanghelich, Kayhan and others},
journal={arXiv preprint arXiv:2201.02831},
year={2022}
}
Download dataset
:modelscope-code[]
许可
[CC BY-SA 4.0]
数据加载示例（图像类）
from PIL import Image
import glob, os
files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
+ glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
+ glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)
# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path) # 要求 子目录=类别
目录组织与标注格式以源站说明和下载后实际文件为准。
数据缺失？
依托陕西、四川两大基地，我们与超过十家三甲医院建立直接合作关系，覆盖合规授权、采集治理、专业标注、数据交付的全流程，为AI医疗团队提供即拿即用的高质量临床数据。
⚡️ 需要数据支持或标注服务？立即联系我们获取专业方案。
