Skip to content
Open Imaging Index
MRI Ultrasound Brain

ReMIND

The Brain Resection Multimodal Imaging Database (ReMIND)

Preoperative MRI, intraoperative MRI and tracked 3D intraoperative ultrasound of brain tumor patients who had image-guided resection, shared for brain shift, registration and segmentation research. 114 patients from one hospital, with tumor, residual tumor, cerebrum and ventricle segmentations.

Go to dataset Edit Report

Overview

ReMIND collects the imaging acquired around brain tumor surgery in the AMIGO suite of Brigham and Women's Hospital in Boston, where resections were guided by both intraoperative ultrasound and intraoperative MRI. Neuronavigation built on preoperative MRI becomes inaccurate once the brain shifts during surgery, and data that pair preoperative and intraoperative scans of several modalities in the same patient are rare. The collection was released on The Cancer Imaging Archive in 2023 under CC BY 4.0, with a data descriptor in Scientific Data in 2024.

Composition

114 patients operated between November 2018 and August 2022, kept from 123 consecutive cases after 9 were dropped for corrupted or poor-quality images. 92 had gliomas (mostly astrocytoma, glioblastoma and oligodendroglioma), 11 had metastases and 11 other lesions. Ages at surgery range from 20 to 76 years; 61 patients are male and 53 female. 51 operations were reoperations. The release has 369 preoperative MRI series, 301 intraoperative MRI series and 320 3D ultrasound volumes, plus a clinical table with histopathology, WHO grade, IDH, 1p/19q and MGMT status.

Acquisition

Preoperative T1, contrast-enhanced T1, T2 and T2-FLAIR come from many scanners and sites (mostly Siemens at 3 T, with 1.5 T, 7 T and 1.16 T scans), rigidly co-registered in Brainlab Elements. Intraoperative MRI was acquired on a 3 T wide-bore Siemens Magnetom Verio. Ultrasound came from a tracked BK5000 system with an N13C5 curvilinear probe swept across the craniotomy and reconstructed into 3D volumes, aimed at three time points: before dural opening, after dural opening, and just before intraoperative MRI. MR scans were defaced. Images are DICOM; segmentations are DICOM SEG and NRRD.

Annotations

Manual preoperative whole tumor (113 cases), tumor target (3), prior resection cavity (21) and intraoperative residual tumor (58) segmentations by neurosurgeons, plus automatic Brainlab cerebrum (89) and ventricle (54) masks.

Known limitations

  • Single institution; preoperative protocols vary between scanners and sites.
  • Cerebrum and ventricle masks are automatic and not meant as a segmentation gold standard.
  • 22 cases miss ultrasound at one or more time points.
  • MRI-ultrasound landmarks were not yet released with the collection.

Cohort

Aggregate numbers from the sources below. Bars are relative to the 114 subjects.

Sex

  • Female 53 46%
  • Male 61 54%

Table 2 in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

Modality

  • MRI 114 100%
  • Ultrasound 114 100%

Abstract in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

Condition

Groups can overlap

  • Glioma 92 81%
  • Astrocytoma 33 29%
  • Glioblastoma 31 27%
  • Oligodendroglioma 24 21%
  • Brain metastases 11 10%
  • Meningioma 1 <1%

Abstract, Table 2 in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

Age

mean 46.8 ± 14.8, median 47.5, range 20 to 76

0
10
20
27
20-2930-3940-4950-5960-6970-79

Age in ReMIND clinical data table, September 2023 (CC BY 4.0, open download); Table 2, Methods, Clinical, demographic, and pathology data in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

Contrast / sequence

Groups can overlap

  • T1-weighted, contrast enhanced 105 92%

Table 3 in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

Field strength

  • 3 T 81 71%
  • 1.5 T 24 21%
  • 7 T 2 2%
  • 1.16 T 2 2%

Table 2 in Juvekar et al. 2024, ReMIND, Scientific Data 11, 494 (preprint medRxiv 10.1101/2023.09.14.23295596)

License and access

Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer

Access
Open download

Download without an account

DICOM images and DICOM SEG segmentations via the TCIA Data Retriever, NRRD segmentations via IBM Aspera, and the clinical table as a direct download. The collection is also listed in the NCI Imaging Data Commons.

Access page

Creative Commons Attribution 4.0 International

Use, share and adapt the data for any purpose, including commercial use, as long as you credit the creators.

Every item on the TCIA collection page (images and segmentations, NRRD segmentations, clinical data) is listed under CC BY 4.0. TCIA also asks users to follow its Data Usage Policy and to cite the dataset DOI.

Original license text Version read: 4.0 Checked 2026-10-07

What you can do

  • Yes
  • Yes
  • Yes
  • Yes
  • Yes

What you can share

  • Yes
  • Yes
  • Yes

What you must do

  • Yes
  • Share alike No
  • No
  • No
  • Manuscript review No
  • Release code No
  • Return results No
  • Delete after use No

Limits

  • No
  • Location limits No

Citation

Juvekar P, Dorent R, Kögl F, et al. ReMIND: The Brain Resection Multimodal Imaging Database. Scientific Data 11, 494 (2024). https://doi.org/10.1038/s41597-024-03295-z. Data: Juvekar P, Dorent R, Kögl F, et al. (2023). The Brain Resection Multimodal Imaging Database (ReMIND) (Version 1) [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/3RAG-D070

Sources

Every number on this page comes from one of these documents. Each chart names the table or page it is taken from. The raw numbers are in stats.csv.