Open access or shared research data must comply with (HIPAA) patient privacy regulations. These regulations require the de-identification of datasets before they can be placed in the public domain. The process of image de-identification is time consuming, requires significant human resources, and is prone to human error. Automated image de-identification algorithms have been developed but the research community requires some method of evaluation before such tools can be widely accepted. This evaluation requires a robust dataset that can be used as part of an evaluation process for de-identification algorithms.

We developed a DICOM dataset that can be used to evaluate the performance of de-identification algorithms. DICOM image information objects were selected from datasets published in TCIA. Synthetic Protected Health Information (PHI) was generated and inserted into selected DICOM data elements to mimic typical clinical imaging exams. The evaluation dataset was de-identified by a TCIA curation team using standard TCIA tools and procedures. We are publishing the evaluation dataset (containing synthetic PHI) and de-identified evaluation dataset (result of TCIA curation) in advance of a potential competition, sponsored by the National Cancer Institute (NCI), for de-identification algorithm evaluation, and de-identification of medical image datasets. The evaluation dataset published here is a subset of a larger evaluation dataset that was created under contract for the National Cancer Institute. This subset is being published to allow researchers to test their de-identification algorithms and promote standardized procedures for validating automated de-identification.

This dataset contains a collection of medical imaging files for use in the "Medical Image Processing with Python" lesson, developed by the Netherlands eScience Center.

The dataset includes:

1. SimpleITK compatible files: MRI T1 and CT scans (training_001_mr_T1.mha, training_001_ct.mha), digital X-ray (digital_xray.dcm in DICOM format), neuroimaging data (A1_grayT1.nrrd, A1_grayT2.nrrd). Data have been downloaded from here.
2. MRI data: a T2-weighted image (OBJECT_phantom_T2W_TSE_Cor_14_1.nii in NIfTI-1 format). Data have been downloaded from here.
3. Example images for the machine learning lesson: chest X-rays (rotatechest.png, other_op.png), cardiomegaly example (cardiomegaly_cc0.png).
4. Additional anonymized data: TBA

These files represent various medical imaging modalities and formats commonly used in clinical research and practice. They are intended for educational purposes, allowing students to practice image processing techniques, machine learning applications, and statistical analysis of medical images using Python libraries such as scikit-image, pydicom, and SimpleITK.

The Lung Image Database Consortium image collection (LIDC-IDRI) consists of diagnostic and lung cancer screening thoracic computed tomography (CT) scans with marked-up annotated lesions. It is a web-accessible international resource for development, training, and evaluation of computer-assisted diagnostic (CAD) methods for lung cancer detection and diagnosis. Initiated by the National Cancer Institute (NCI), further advanced by the Foundation for the National Institutes of Health (FNIH), and accompanied by the Food and Drug Administration (FDA) through active participation, this public-private partnership demonstrates the success of a consortium founded on a consensus-based process.

Seven academic centers and eight medical imaging companies collaborated to create this data set which contains 1018 cases. Each subject includes images from a clinical thoracic CT scan and an associated XML file that records the results of a two-phase image annotation process performed by four experienced thoracic radiologists. In the initial blinded-read phase, each radiologist independently reviewed each CT scan and marked lesions belonging to one of three categories ("nodule > or =3 mm," "nodule <3 mm," and "non-nodule > or =3 mm"). In the subsequent unblinded-read phase, each radiologist independently reviewed their own marks along with the anonymized marks of the three other radiologists to render a final opinion. The goal of this process was to identify as completely as possible all lung nodules in each CT scan without requiring forced consensus.

Note : The TCIA team strongly encourages users to review pylidc and the Standardized representation of the TCIA LIDC-IDRI annotations using DICOM (DICOM-LIDC-IDRI-Nodules) of the annotations/segmentations included in this dataset before developing custom tools to analyze the XML version.

Abstract

These resources comprise a large and diverse collection of multi-site, multi-modality, and multi-cancer clinical DICOM images from 538 subjects infused with synthetic PHI/PII in areas encountered by TCIA curation teams. Also provided is a TCIA-curated version of the synthetic dataset, along with mapping files for mapping identifiers between the two.

This new MIDI data resource includes DICOM datasets used in the Medical Image De-Identification Benchmark (MIDI-B) challenge at MICCAI 2024. They are accompanied by ground truth answer keys and a validation script for evaluating the effectiveness of medical image de-identification workflows. The validation script systematically assesses de-identified data against an answer key outlining appropriate actions and values for proper de-identification of medical images, promoting safer and more consistent medical image sharing.

Introduction

Medical imaging research increasingly relies on large-scale data sharing. However, reliable de-identification of DICOM images still presents significant challenges due to the wide variety of DICOM header elements and pixel data where identifiable information may be embedded. To address this, we have developed an openly accessible synthetic dataset containing artificially generated protected health information (PHI) and personally identifiable information (PII).

These resources complement our earlier work (Pseudo-PHI-DICOM-data ) hosted on The Cancer Imaging Archive. As an example of its use, we also provide a version curated by The Cancer Imaging Archive (TCIA) curation team. This resource builds upon best practices emphasized by the MIDI Task Group who underscore the importance of transparency, documentation, and reproducibility in de-identification workflows, part of the themes at recent conferences (Synapse:syn53065760) and workshops (2024 MIDI-B Challenge Workshop).

This framework enables objective benchmarking of de-identification performance, promotes transparency in compliance with regulatory standards, and supports the establishment of consistent best practices for sharing clinical imaging data. We encourage the research community to use these resources to enhance and standardize their medical image de-identification workflows.

Methods

Subject Inclusion and Exclusion Criteria

The source data were selected from imaging already hosted in de-identified form on TCIA. Imaging containing faces were excluded, and no new human studies were performed for his project.

Data Acquisition

To build the synthetic dataset, image series were selected from TCIA’s curated datasets to represent a broad range of imaging modalities (CR, CT, DX, MG, MR, PT, SR, US) , manufacturers including (GE, Siemens, Varian , Confirma, Agfa, Eigen, Elekta, Hologic, KONICA MINOLTA, others) , scan parameters, and regions of the body. These were processed to inject the synthetic PHI/PII as described.

Data Analysis

Synthetic pools of PHI, like subject and scanning institution information, were generated using the Python package Faker (https://pypi.org/project/Faker/8.10.3/). These were inserted into DICOM metadata of selected imaging files using a system of inheritable rule-based templates outlining re-identification functions for data insertion and logging for answer key creation. Text was also burned-in to the pixel data of a number of images. By systematically embedding realistic synthetic PHI into image headers and pixel data, accompanied by a detailed ground-truth answer key, our framework enables users transparency, documentation, and reproducibility in de-identification practices, aligned with the HIPAA Safe Harbor method, DICOM PS3.15 Confidentiality Profiles, and TCIA best practices.

Usage Notes

This DICOM collection is split into two datasets, synthetic and curated. The synthetic dataset is the PHI/PII infused DICOM collection accompanied by a validation script and answer keys for testing, refining and benchmarking medical image de-identification pipelines. The curated dataset is a version of the synthetic dataset curated and de-identified by members of The Cancer Imaging Archive curation team. It can be used as a guide, an example of medical image curation best practices. For the purposes of the De-Identification challenge at MICCAI 2024, the synthetic and curated datasets each contain two subsets, a portion for Validation and the other for Testing.

To link a curated dataset to the original synthetic dataset and answer keys, a mapping between the unique identifiers (UIDs) and patient IDs must be provided in CSV format to the evaluation software. We include the mapping files associated with the TCIA-curated set as an example. Lastly, for both the Validation and Testing datasets, an answer key in sqlite.db format is provided. These components are for use with the Python validation script linked below (4). Combining these components, a user developing or evaluating de-identification methods can ensure they meet a specification for successfully de-identifying medical image data.

Brain MRI Dataset, Normal Brain Dataset, Anomaly Classification & Detection

The dataset consists of .dcm files containing MRI scans of the brain of the person with a normal brain. The images are labeled by the doctors and accompanied by report in PDF-format. The dataset includes 7 studies, made from the different angles which provide a comprehensive understanding of a normal brain structure and useful in training brain anomaly classification algorithms.

  MRI study angles… See the full description on the dataset page: https://huggingface.co/datasets/TrainingDataPro/dicom-brain-dataset.

The Data Integration & Imaging Informatics (DI-Cubed) project explored the issue of lack of standardized data capture at the point of data creation, as reflected in the non-image data accompanying various TCIA breast cancer collections. The work addressed the desire for semantic interoperability between various NCI initiatives by aligning on common clinical metadata elements and supporting use cases that connect clinical, imaging, and genomics data. Accordingly, clinical and measurement data was imported into I2B2 and cross-mapped to industry standard concepts for names and values including those derived from BRIDG, CDISC SDTM, DICOM Structured Reporting models and using NCI Thesaurus, SNOMED CT and LOINC controlled terminology. A subset of the standardized data was then exported from I2B2 to CSV and thence converted to DICOM SR according to the the DICOM Breast Imaging Report template [1] , which supports description of patient characteristics, histopathology, receptor status and clinical findings including measurements. The purpose was not to advocate DICOM SR as an appropriate format for interchange or storage of such information for query purposes, but rather to demonstrate that use of standard concepts harmonized across multiple collections could be transformed into an existing standard report representation. The DICOM SR can be stored and used together with the images in repositories such as TCIA and in image viewers that support rendering of DICOM SR content. During the project, various deficiencies in the DICOM Breast Imaging Report template were identified with respect to describing breast MR studies, laterality of findings versus procedures, more recently developed receptor types, and patient characteristics and status. These were addressed via DICOM CP 1838, finalized in Jan 2019, and this subset reflects those changes. DICOM Breast Imaging Report Templates available from: http://dicom.nema.org/medical/dicom/current/output/chtml/part16/sect_BreastImagingReportTemplates.html

The Magnetic Resonance - Computed Tomography (MR-CT) Jordan University Hospital (JUH) dataset has been collected after receiving Institutional Review Board (IRB) approval of the hospital and consent forms have been obtained from all patients. All procedures has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).

The dataset consists of 2D image slices extracted using the RadiAnt DICOM viewer software. The extracted images are transformed to DICOM image data format with a resolution of 256x256 pixels. There are a total of 179 2D axial image slices referring to 20 patient volumes (90 MR and 89 CT 2D axial image slices). The dataset contains MR and CT brain tumour images with corresponding segmentation masks. The MR images of each patient were acquired with a 5.00mm T Siemens Verio 3T using a T2-weighted without contrast agent, 3 Fat sat pulses (FS), 2500-4000 TR, 20-30 TE, and 90/180 flip angle. The CT images were acquired with Siemens Somatom scanner with 2.46mGY.cm dose length, 130KV voltage, 113-327 mAs tube current, topogram acquisition protocol, 64 dual source, one projection, and slice thickness of 7.0mm. Smooth and sharp filters have been applied to the CT images. The MR scans have a resolution of 0.7x0.6x5 mm^3, while the CT scans have a resolution of 0.6x0.6x7 mm^3.

More information and the application of the dataset can be found in the following research paper:

Alaa Abu-Srhan; Israa Almallahi; Mohammad Abushariah; Waleed Mahafza; Omar S. Al-Kadi. Paired-Unpaired Unsupervised Attention Guided GAN with Transfer Learning for Bidirectional Brain MR-CT Synthesis. Comput. Biol. Med. 136, 2021. doi: https://doi.org/10.1016/j.compbiomed.2021.104763.

This dataset consists of CT and PET-CT DICOM images of lung cancer subjects with XML Annotation files that indicate tumor location with bounding boxes. The images were retrospectively acquired from patients with suspicion of lung cancer, and who underwent standard-of-care lung biopsy and PET/CT. Subjects were grouped according to a tissue histopathological diagnosis. Patients with Names/IDs containing the letter 'A' were diagnosed with Adenocarcinoma, 'B' with Small Cell Carcinoma, 'E' with Large Cell Carcinoma, and 'G' with Squamous Cell Carcinoma.

The images were analyzed on the mediastinum (window width, 350 HU; level, 40 HU) and lung (window width, 1,400 HU; level, –700 HU) settings. The reconstructions were made in 2mm-slice-thick and lung settings. The CT slice interval varies from 0.625 mm to 5 mm. Scanning mode includes plain, contrast and 3D reconstruction.

Before the examination, the patient underwent fasting for at least 6 hours, and the blood glucose of each patient was less than 11 mmol/L. Whole-body emission scans were acquired 60 minutes after the intravenous injection of 18F-FDG (4.44MBq/kg, 0.12mCi/kg), with patients in the supine position in the PET scanner. FDG doses and uptake times were 168.72-468.79MBq (295.8±64.8MBq) and 27-171min (70.4±24.9 minutes), respectively. 18F-FDG with a radiochemical purity of 95% was provided. Patients were allowed to breathe normally during PET and CT acquisitions. Attenuation correction of PET images was performed using CT data with the hybrid segmentation method. Attenuation corrections were performed using a CT protocol (180mAs,120kV,1.0pitch). Each study comprised one CT volume, one PET volume and fused PET and CT images: the CT resolution was 512 × 512 pixels at 1mm × 1mm, the PET resolution was 200 × 200 pixels at 4.07mm × 4.07mm, with a slice thickness and an interslice distance of 1mm. Both volumes were reconstructed with the same number of slices. Three-dimensional (3D) emission and transmission scanning were acquired from the base of the skull to mid femur. The PET images were reconstructed via the TrueX TOF method with a slice thickness of 1mm.

The location of each tumor was annotated by five academic thoracic radiologists with expertise in lung cancer to make this dataset a useful tool and resource for developing algorithms for medical diagnosis. Two of the radiologists had more than 15 years of experience and the others had more than 5 years of experience. After one of the radiologists labeled each subject the other four radiologists performed a verification, resulting in all five radiologists reviewing each annotation file in the dataset. Annotations were captured using Labellmg. The image annotations are saved as XML files in PASCAL VOC format, which can be parsed using the PASCAL Development Toolkit: https://pypi.org/project/pascal-voc-tools/. Python code to visualize the annotation boxes on top of the DICOM images can be downloaded here.

Two deep learning researchers used the images and the corresponding annotation files to train several well-known detection models which resulted in a maximum a posteriori probability (MAP) of around 0.87 on the validation set.

training

The Dataset is a meticulously curated high-quality dataset specifically designed for semantic-guided image fusion in the medical domain. This dataset aims to facilitate advanced research and development in multimodal medical image analysis by providing a comprehensive collection of images from various imaging modalities.

This dataset corresponds to a collection of images and/or image-derived data available from National Cancer Institute Imaging Data Commons (IDC) [1]. This dataset was converted into DICOM representation and ingested by the IDC team. You can explore and visualize the corresponding images using IDC Portal here: NLM-Visible-Human-Project. You can use the manifests included in this Zenodo record to download the content of the collection following the Download instructions below.

Collection description

The NLM Visible Human Project [2] has created publicly-available complete, anatomically detailed, three-dimensional representations of a human male body and a human female body. Specifically, the VHP provides a public-domain library of cross-sectional cryosection, CT, and MRI images obtained from one male cadaver and one female cadaver. The Visible Man data set was publicly released in 1994 and the Visible Woman in 1995.

The data sets were designed to serve as (1) a reference for the study of human anatomy, (2) public-domain data for testing medical imaging algorithms, and (3) a test bed and model for the construction of network-accessible image libraries. The VHP data sets have been applied to a wide range of educational, diagnostic, treatment planning, virtual reality, artistic, mathematical, and industrial uses. About 4,000 licensees from 66 countries were authorized to access the datasets. As of 2019, a license is no longer required to access the VHP datasets.

Courtesy of the U.S. National Library of Medicine. Release of this collection by IDC does not indicate or imply that NLM has endorsed its products/services/applications. Please see the Visible Human Project information page to learn more about the images and to obtain any supporting metadata for this collection. Note that this collection may not reflect the most current/accurate data available from NLM.

Citation guidelines can be found on the National Library of Medicine Terms and Conditions information page.

Files included

A manifest file's name indicates the IDC data release in which a version of collection data was first introduced. For example, collection_id-idc_v8-aws.s5cmd corresponds to the contents of the collection_id collection introduced in IDC data release v8. If there is a subsequent version of this Zenodo page, it will indicate when a subsequent version of the corresponding collection was introduced.

1. nlm_visible_human_project-idc_v15-aws.s5cmd: manifest of files available for download from public IDC Amazon Web Services buckets
2. nlm_visible_human_project-idc_v15-gcs.s5cmd: manifest of files available for download from public IDC Google Cloud Storage buckets
3. nlm_visible_human_project-idc_v15-dcf.dcf: Gen3 manifest (for details see https://learn.canceridc.dev/data/organization-of-data/guids-and-uuids)

Note that manifest files that end in -aws.s5cmd reference files stored in Amazon Web Services (AWS) buckets, while -gcs.s5cmd reference files in Google Cloud Storage. The actual files are identical and are mirrored between AWS and GCP.

Download instructions

Each of the manifests include instructions in the header on how to download the included files.

To download the files using .s5cmd manifests:

1. install idc-index package: pip install --upgrade idc-index
2. download the files referenced by manifests included in this dataset by passing the .s5cmd manifest file: idc download manifest.s5cmd.

To download the files using .dcf manifest, see manifest header.

Acknowledgments

Imaging Data Commons team has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Task Order No. HHSN26110071 under Contract No. HHSN261201500003l.

References

[1] Fedorov, A., Longabaugh, W. J. R., Pot, D., Clunie, D. A., Pieper, S. D., Gibbs, D. L., Bridge, C., Herrmann, M. D., Homeyer, A., Lewis, R., Aerts, H. J. W., Krishnaswamy, D., Thiriveedhi, V. K., Ciausu, C., Schacherer, D. P., Bontempi, D., Pihl, T., Wagner, U., Farahani, K., Kim, E. & Kikinis, R. National Cancer Institute Imaging Data Commons: Toward Transparency, Reproducibility, and Scalability in Imaging Artificial Intelligence. RadioGraphics (2023). https://doi.org/10.1148/rg.230180

[2] Spitzer, V., Ackerman, M. J., Scherzinger, A. L. & Whitlock, D. The visible human male: a technical report. J. Am. Med. Inform. Assoc. 3, 118–130 (1996). https://doi.org/10.1136/jamia.1996.96236280

This dataset contains key characteristics about the data described in the Data Descriptor A DICOM dataset for evaluation of medical image de-identification. Contents:

    1. human readable metadata summary table in CSV format


    2. machine readable metadata file in JSON format

Medical Image

## Overview

Medical Image is a dataset for object detection tasks - it contains Mask Glove Vial Syringe Spluit I annotations for 2,908 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Sparsity of annotated data is a major limitation in medical image processing tasks such as registration. Registered multimodal image data are essential for the diagnosis of medical conditions and the success of interventional medical procedures. To overcome the shortage of data, we present a method that allows the generation of annotated multimodal 4D datasets. We use a CycleGAN network architecture to generate multimodal synthetic data from the 4D extended cardiac–torso (XCAT) phantom and real patient data. Organ masks are provided by the XCAT phantom; therefore, the generated dataset can serve as ground truth for image segmentation and registration. Compared to real patient data, the synthetic data showed good agreement regarding the image voxel intensity distribution and the noise characteristics. The generated T1-weighted magnetic resonance imaging, computed tomography (CT), and cone beam CT images are inherently co-registered.

The dataset contains multi-modal data from over 70,000 open access and de-identified case reports, including metadata, clinical cases, image captions and more than 130,000 images. Images and clinical cases belong to different medical specialties, such as oncology, cardiology, surgery and pathology. The structure of the dataset allows to easily map images with their corresponding article metadata, clinical case, captions and image labels. Details of the data structure can be found in the file data_dictionary.csv.

More than 90,000 patients and 280,000 medical doctors and researchers were involved in the creation of the articles included in this dataset. The citation data of each article can be found in the metadata.parquet file.

Refer to the examples showcased in this GitHub repository to understand how to optimize the use of this dataset.The license of the dataset as a whole is CC BY-NC-SA. However, its individual contents may have less restrictive license types (CC BY, CC BY-NC, CC0). For instance, regarding image filess, 66K of them are CC BY, 32K are CC BY-NC-SA, 32K are CC BY-NC, and 20 of them are CC0.

Radiographs are used as the most critical imaging tool for identifying spine anomalies in clinical practice [1]. The evaluation of spinal bone lesions, however, is a challenging task for radiologists. To the best of our knowledge, no existing studies are devoted to developing and evaluating a comprehensive system for classifying and localizing multiple spine lesions from X-ray scans. The lack of large-scale spine X-ray datasets with high-quality images and human expert annotations is the key obstacle. To fill this gap, we introduce a large-scale annotated medical image dataset for spinal lesion detection and classification from radiographs. The dataset, called VinDr-SpineXR, contains 10,466 spine X-ray images from 5,000 studies, each of which is manually annotated with 13 types of abnormalities by an experienced radiologist with bounding boxes around abnormal findings. This is the largest dataset to date that provides radiologist's bounding-box annotations for developing supervised-learning algorithms for spine X-ray analysis.

Transform healthcare diagnostics with image segmentation. Dive into advanced techniques for detailed medical imaging, aiding patient care.

Three different datasets of vertebrae with corresponding computed tomography (CT) and ultrasound (US) images are presented. In the first dataset, three human patients lumbar vertebrae are presented and the US images are simulated from their CT images. The second dataset includes corresponding CT, US, and simulated US images of a phantom made from post-mortem canine cervical and thoracic vertebrae. The last phantom consists of the CT, US, and simulated US images of a phantom made from a post-mortem lamb lumbar vertebrae. For each of the two latter datasets, we also provide 15 landmark pairs of matching structures between the CT and US images and performed fiducial registration to acquire a silver standard for assessing image registration.

The datasets can be used to test CT-US image registration techniques and to validate techniques that simulate US from CT.

This dataset contributes DICOM-converted annotations to the publicly available National Cancer Institute Imaging Data Commons [1] Prostate-MRI-US-Biopsy collection (https://portal.imaging.datacommons.cancer.gov/explore/filters/?collection_id=Community&collection_id=prostate_mri_us_biopsy). Prostate-MRI-US-Biopsy collection was initially released by The Cancer Imaging Archive (TCIA) [2,3,4]. While the images in this collection are stored in the standard DICOM format, the collection is also accompanied by 1017 semi-automatic segmentations of the prostate and 1317 manual segmentations of target lesions in the STL format. Although STL is a common and practical format for 3D printing, it is not interoperable with many visualization and analysis tools commonly used in medical imaging research and does not provide any standard means to communicate metadata, among other limitations. This dataset contains segmentations of the prostate and target lesions harmonized into DICOM representation. Specifically, we created DICOM Encapsulated 3D Manufacturing Model objects (M3D modality) that includes the original STL content enriched with the DICOM metadata. Furthermore, we created an alternative encoding of the surface segmentations by rasterizing them and saving the result as a DICOM Segmentation object (SEG modality). As a result, the contributed DICOM objects can be stored in any DICOM server that supports those objects (including Google Healthcare DICOM stores), and the DICOM Segmentations can be visualized using off-the-shelf tools, such as OHIF Viewer. Conversion from STL to DICOM M3D modality was performed using PixelMed toolkit (https://www.pixelmed.com/dicomtoolkit.html). Conversion from STL to DICOM SEG was done in 2 steps. We used Slicer (https://www.slicer.org/) to rasterize the surface segmentation to the matrix of the segmented image, which were next converted to DICOM SEGs using dcmqi (https://github.com/QIICR/dcmqi) [5]. Resulting objects were validated using dicom3tools dciodvfy (https://www.dclunie.com/dicom3tools.html). Details describing the conversion process as well as the details on how to access the encapsulated STL content from the DICOM m3D files are provided in this GitHub repository: https://github.com/ImagingDataCommons/prostate_mri_us_biopsy_dcm_conversion. Specific files included in the record are:

Prostate-MRI-US-Biopsy-DICOM-Annotations.zip: DICOM M3D and SEG files, organized into the folder hierarchy following this pattern: Prostate-MRI-US-Biopsy/%PatientID/%StudyInstanceUID/%SeriesNumber-%Modality-%SeriesDescription.dcm referenced_images_sorted-idc_file_manifest.s5cmd: IDC manifest for downloading the T2W MRI images corresponding to the annotations. To download the files in this manifest, first install s5cmd (https://github.com/peak/s5cmd), and run the following command: s5cmd --no-sign-request --endpoint-url https://s3.amazonaws.com run referenced_images_sorted-idc_file_manifest.s5cmd. Files will be organized in the Prostate-MRI-US-Biopsy/%PatientID/%StudyInstanceUID/ folder hierarchy upon download. References [1] Fedorov, A., Longabaugh, W. J. R., Pot, D., Clunie, D. A., Pieper, S., Aerts, H. J. W. L., Homeyer, A., Lewis, R., Akbarzadeh, A., Bontempi, D., Clifford, W., Herrmann, M. D., Höfener, H., Octaviano, I., Osborne, C., Paquette, S., Petts, J., Punzo, D., Reyes, M., Schacherer, D. P., Tian, M., White, G., Ziegler, E., Shmulevich, I., Pihl, T., Wagner, U., Farahani, K. & Kikinis, R. NCI Imaging Data Commons. Cancer Res. 81, 4188–4193 (2021). doi: 10.1158/0008-5472.CAN-21-0950. [2] Natarajan, S., Priester, A., Margolis, D., Huang, J., & Marks, L. (2020). Prostate MRI and Ultrasound With Pathology and Coordinates of Tracked Biopsy (Prostate-MRI-US-Biopsy) (version 2) [Data set]. The Cancer Imaging Archive. DOI: 10.7937/TCIA.2020.A61IOC1A [3] Sonn GA, Natarajan S, Margolis DJ, MacAiran M, Lieu P, Huang J, Dorey FJ, Marks LS. Targeted biopsy in the detection of prostate cancer using an office based magnetic resonance ultrasound fusion device. Journal of Urology 189, no. 1 (2013): 86-91. DOI: 10.1016/j.juro.2012.08.095 [4] Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, Moore S, Phillips S, Maffitt D, Pringle M, Tarbox L, Prior F. The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information Repository, Journal of Digital Imaging, Volume 26, Number 6, December, 2013, pp 1045-1057. DOI: 10.1007/s10278-013-9622-7 [5] Herz, C., Fillion-Robin, J.-C., Onken, M., Riesmeier, J., Lasso, A., Pinter, C., Fichtinger, G., Pieper, S., Clunie, D., Kikinis, R. & Fedorov, A. dcmqi: An Open Source Library for Standardized Communication of Quantitative Image Analysis Results Using DICOM. Cancer Res. 77, e87–e90 (2017). DOI: 10.1158/0008-5472.CAN-17-0336.

This CBIS-DDSM (Curated Breast Imaging Subset of DDSM) is an updated and standardized version of the Digital Database for Screening Mammography (DDSM). The DDSM is a database of 2,620 scanned film mammography studies. It contains normal, benign, and malignant cases with verified pathology information. The scale of the database along with ground truth validation makes the DDSM a useful tool in the development and testing of decision support systems. The CBIS-DDSM collection includes a subset of the DDSM data selected and curated by a trained mammographer. The images have been decompressed and converted to DICOM format. Updated ROI segmentation and bounding boxes, and pathologic diagnosis for training data are also included. A manuscript describing how to use this dataset in detail is available at https://www.nature.com/articles/sdata2017177.

Published research results from work in developing decision support systems in mammography are difficult to replicate due to the lack of a standard evaluation data set; most computer-aided diagnosis (CADx) and detection (CADe) algorithms for breast cancer in mammography are evaluated on private data sets or on unspecified subsets of public databases. Few well-curated public datasets have been provided for the mammography community. These include the DDSM, the Mammographic Imaging Analysis Society (MIAS) database, and the Image Retrieval in Medical Applications (IRMA) project. Although these public data sets are useful, they are limited in terms of data set size and accessibility.

For example, most researchers using the DDSM do not leverage all its images for a variety of historical reasons. When the database was released in 1997, computational resources to process hundreds or thousands of images were not widely available. Additionally, the DDSM images are saved in non-standard compression files that require the use of decompression code that has not been updated or maintained for modern computers. Finally, the ROI annotations for the abnormalities in the DDSM were provided to indicate a general position of lesions, but not a precise segmentation for them. Therefore, many researchers must implement segmentation algorithms for accurate feature extraction. This causes an inability to directly compare the performance of methods or to replicate prior results. The CBIS-DDSM collection addresses that challenge by publicly releasing an curated and standardized version of the DDSM for evaluation of future CADx and CADe systems (sometimes referred to generally as CAD) research in mammography.

Please note that the image data for this collection is structured such that each participant has multiple patient IDs. For example, participant 00038 has 10 separate patient IDs which provide information about the scans within the IDs (e.g. Calc-Test_P_00038_LEFT_CC, Calc-Test_P_00038_RIGHT_CC_1). This makes it appear as though there are 6,671 patients according to the DICOM metadata, but there are only 1,566 actual participants in the cohort.

For scientific and other inquiries about this dataset, please contact TCIA's Helpdesk.