EzBIDS: The open cloud service for automated, validated DICOM to BIDS conversion

Presented During:

Poster No:


Submission Type:

Abstract Submission 


Daniel Levitas1, Soichi Hayashi1, Franco Pestilli1


1Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN

First Author:

Daniel Levitas  
Department of Psychological and Brain Sciences, Indiana University
Bloomington, IN


Soichi Hayashi  
Department of Psychological and Brain Sciences, Indiana University
Bloomington, IN
Franco Pestilli  
Department of Psychological and Brain Sciences, Indiana University
Bloomington, IN


Over the past several years there has been a concerted effort within the neuroimaging field to organize and standardize imaging data according to the specifications laid out in the Brain Imaging Dataset Structure standard (BIDS; Gorgolewski et al., 2016). Adhering to this standard is of great benefit for data sharing and replication of previous studies. Yet, as of today data BIDSification process is nontrivial and requires a considerable amount of time as well as advanced software skills. Currently, a few dozen open-source projects have developed code to allow convert DICOM to BIDS. However, all these projects require the use of a Linux terminal and programming; no tool currently exists targeting the broader community of potential BIDS standards users with a diverse background spanning from no to limited coding skills. To broaden the reach and adoption of the BIDS standard more mechanisms are necessary that can lower the barrier of entry to data standardization. We present a new cloud computing service that automatically maps DICOM files to the BIDS standard. The service is publicly available at brainlife.io/ezbids but can also be deployed on other resources.


EzBIDS is a web-based hosted DICOM to BIDS conversion service available at brainlife.io/ezbids (Avesani et al., Nature Scientific Data 2019). EzBIDS requires neither installation nor Linux terminal knowledge. The cloud service-based model of EzBIDS allows researchers to upload their DICOM datasets to be automatically organized into BIDS-compliant datasets. This service takes advantage of various command-line based tools that currently exist, as well as our new machine learning (ML) modules to drive interactive conversion steps through the web UI. The ML part of the service provides initial suggestions to the users. The UI allows users to correct any mistakes or fill-in missing information, such as subject and session IDs, and task name(s), if this information cannot be discerned from the DICOM headers or file structure/naming. Any correction entered by users is recorded and used to improve the algorithm or ML modules over time. Once this information has been verified, EzBIDS will perform a first-pass to determine which acquisitions should be converted to BIDS and which should not, based on a custom heuristic. Acquisitions that are deemed unconvertible will be flagged for review, with the rationale provided to users to allow them to make an executive decision as to whether or not to convert the acquisition(s) in question. Once the user had made the final checks, EzBIDS incorporates the user feedback to perform the BIDS conversion. In order to broaden its applicability, EzBIDS is validated on imaging acquisition from the major MR vendors (Siemens, GE, and Phillips).
Supporting Image: OHBM_EzBIDS_Fig1.png
   ·Figure 1. EzBIDS schematic workflow


An example of the web user interface is provided below (Figure 2).
Supporting Image: OHBM_MockUI_Fig2.png
   ·Figure 2. Web User Interface Example


EzBIDS is a new generation cloud service for data standardization that combines ML and citizen science methods to advance the neuroimaging field in the efforts to promote FAIR principles. The unique design we propose combining human and ML cooperation will make EzBIDS robust to the passing of time, advance to the BIDS standard definition, as well as to the changes in the interpretation of the DICOM file format by the major magnetic system vendors.

Neuroinformatics and Data Sharing:

Databasing and Data Sharing 1
Workflows 2


Data Registration
Machine Learning
Other - BIDS Conversion; Web User Interface; Open-Source

1|2Indicates the priority used for review

My abstract is being submitted as a Software Demonstration.


Please indicate below if your study was a "resting state" or "task-activation” study.


Healthy subjects only or patients (note that patient studies may also involve healthy subjects):

Healthy subjects

Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.


Was any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.

Not applicable

Was any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.

Not applicable

Please indicate which methods were used in your research:

Other, Please specify  -   Web User Interface Development

Provide references using author date format

Avesani, P., McPherson, B., Hayashi, S., Caiafa, C. F., Henschel, R., Garyfallidis, E., ... & Sporns, O. (2019). The open diffusion data derivatives, brain data upcycling via integrated publishing of derivatives and reproducible open cloud services. Scientific data, 6(1), 69.

Gorgolewski, K. J., Auer, T., Calhoun, V. D., Craddock, R. C., Das, S., Duff, E. P., ... & Handwerker, D. A. (2016). The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments. Scientific Data, 3, 160044.