Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Feb 29;11(1):254.
doi: 10.1038/s41597-024-03021-9.

A large open access dataset of brain metastasis 3D segmentations on MRI with clinical and imaging information

Divya Ramakrishnan  1 , Leon Jekel  2   3 , Saahil Chadha  2 , Anastasia Janas  2   4 , Harrison Moy  2   5 , Nazanin Maleki  2 , Matthew Sala  2   6 , Manpreet Kaur  2   7 , Gabriel Cassinelli Petersen  2   8 , Sara Merkaj  2   9 , Marc von Reppert  2   10 , Ujjwal Baid  11   12 , Spyridon Bakas  11   12 , Claudia Kirsch  2   13   14 , Melissa Davis  2 , Khaled Bousabarah  15 , Wolfgang Holler  15 , MingDe Lin  2   16 , Malte Westerhoff  15 , Sanjay Aneja  17   18 , Fatima Memon  2 , Mariam S Aboian  2
Affiliations

A large open access dataset of brain metastasis 3D segmentations on MRI with clinical and imaging information

Divya Ramakrishnan et al. Sci Data. .

Abstract

Resection and whole brain radiotherapy (WBRT) are standard treatments for brain metastases (BM) but are associated with cognitive side effects. Stereotactic radiosurgery (SRS) uses a targeted approach with less side effects than WBRT. SRS requires precise identification and delineation of BM. While artificial intelligence (AI) algorithms have been developed for this, their clinical adoption is limited due to poor model performance in the clinical setting. The limitations of algorithms are often due to the quality of datasets used for training the AI network. The purpose of this study was to create a large, heterogenous, annotated BM dataset for training and validation of AI models. We present a BM dataset of 200 patients with pretreatment T1, T1 post-contrast, T2, and FLAIR MR images. The dataset includes contrast-enhancing and necrotic 3D segmentations on T1 post-contrast and peritumoral edema 3D segmentations on FLAIR. Our dataset contains 975 contrast-enhancing lesions, many of which are sub centimeter, along with clinical and imaging information. We used a streamlined approach to database-building through a PACS-integrated segmentation workflow.

PubMed Disclaimer

Conflict of interest statement

C.K. – receives royalties from Primal Pictures 3D Informa, has grant funding from the NIH, and has received the Core Curriculum grant from the American Society of Head and Neck Radiology, all unrelated to this work. K.B. – employee of Visage Imaging GmbH. W.H. – employee and stockholder of Visage Imaging GmbH. M.L. – employee and stockholder of Visage Imaging, Inc., and unrelated to this work, receives funding from NIH/NCI R01 CA206180 and NIH/NCI R01 CA275188. M.W. – employee and stockholder of Visage Imaging GmbH. M.S.A. – has collaborations with Visage Imaging, Inc., Blue Earth Diagnostics, Telix, and AAA. She also has a KL2 TR00186 grant from the NCATS foundation. The remaining co-authors do not have any competing interests.

Figures

Fig. 1
Fig. 1
Research PACS annotation layout. The T1, T1 post-contrast, FLAIR, and T2 sequences for one patient are displayed on the eight-viewer layout after alignment with the auto-align tool. The PACS interface incorporates a 3D volumetric tool (white circle/rectangle) and displays labeled segmentations for two brain metastases in the display window (red rectangle).
Fig. 2
Fig. 2
PACS-based segmentations and NIfTI masks for one patient. After auto-alignment of FLAIR and T1 post-contrast sequences, segmentation of whole tumor (“Whole2_FLAIR”), including peritumoral edema, was performed on FLAIR (a), and segmentations of contrast-enhancing lesion (“Core2_PGGE”) and corresponding necrotic portions (“Necrosis2_PGGE”) were performed on T1 post-contrast (b). (c) The combined segmentation masks are shown overlaid on the FLAIR sequence in NIfTI format. The green region represents peritumoral edema, the blue region represents contrast-enhancing tumor, and the red region represents necrotic tumor.
See this image and copyright information in PMC

Update of

References

    1. Kotecha R, Gondi V, Ahluwalia MS, Brastianos PK, Mehta MP. Recent advances in managing brain metastasis. F1000Res. 2018;7:F1000 Faculty Rev–1772. doi: 10.12688/f1000research.15903.1. - DOI - PMC - PubMed
    1. Boire A, Brastianos PK, Garzia L, Valiente M. Brain metastasis. Nat Rev Cancer. 2020;20:4–11. doi: 10.1038/s41568-019-0220-y. - DOI - PubMed
    1. Buecker R, et al. Risk factors to identify patients who may not benefit from whole brain irradiation for brain metastases - a single institution analysis. Radiation Oncology. 2019;14:41. doi: 10.1186/s13014-019-1245-9. - DOI - PMC - PubMed
    1. Park YW, et al. Differentiation of recurrent glioblastoma from radiation necrosis using diffusion radiomics with machine learning model development and external validation. Sci Rep. 2021;11:2913. doi: 10.1038/s41598-021-82467-y. - DOI - PMC - PubMed
    1. Xue J, et al. Biological implications of whole-brain radiotherapy versus stereotactic radiosurgery of multiple brain metastases. J Neurosurg. 2014;121(Suppl):60–68. doi: 10.3171/2014.7.GKS141229. - DOI - PubMed