CLINT (CLInical Neurosynth Translator)

An electronic medical record (EMR) pipeline using neurosynth to link clinical data with fMRI research literature

Hackathon team: Lead: Dylan M. Nielson - SysAdmins: Ben Cohen, Meghna Verma, Zehai Wang - Writers: Mostafa Lotfi, Ben Cohen, Dylan M. Nielson, Meghna Verma, Zehai Wang

How to cite this work in a publication: (doi: )

CLINT is a data gathering and query pipeline that will parse electronic medical record (EMR) reports, interface with Neurosynth to produce a list of symptoms correlated to structures or structures correlated to symptoms queried and return a report in an EMR ingestible format. Neurosynth is a database of functional neuroimaging results linking function with brain locations from over 11,000 published functional magnetic resonance imaging (fMRI) studies.

This project was part of the Spring 2018 NCBI Hackathon

Application 1

Clinical radiology reports indicate the structures involved in any clinical finding, but might not mak$

Application 2

Similarly, sometime it takes a long time to get a patient into the scanner, if you know what symptoms $

Application 3

When searching for patients to enroll in clinical trials study directors could provide a list of desired patient symptoms and CLINT could output the corresponding anatomical mask and EMRs

In a perfect world here's how CLINT might work:

Dependencies:computer:

Software: Neurosynth Docker

Databases

EMR's:OHDSI/ETL-CMS github Brain Atlas: FMRIB Software Library Main page FMRIB Software Library Atlases FSL-Harvard-Oxford-Atlases

Workflow Implementation

In order to implement CLINT follow the steps below:

Step 1.

Sample Outputs

Todo:

1. Parse reports
2. Query Neurosynth
3. Aggregate Neurosynth reports
4. Create FHIR JSON
5. Containerize endpoint and drop it on an AWS box

Acknowledgements

MetagenomicsAntibioticResistance github used as a template for formatting and organization