LesionBank.org: An Open Source Platform for Brain Lesion Case Reports

Poster No:

2240 

Submission Type:

Abstract Submission 

Authors:

Michael Ferguson1, Vicky Chen2, Joseph Turner3, Anish Suvarna4, Eli Baughan5, Kiana Bunnell5, Jared Nielsen5, Frederic Schaper6

Institutions:

1Harvard Medical School, Boston, MA, 2Massachusetts Institute of Technology, Cambridge, MA, 3Brigham and Women's Hospital, Boston, MA, 4Thomas Jefferson High School for Science and Technology, Alexandria, VA, 5Brigham Young University, Provo, UT, 6Brigham and Women’s Hospital, Harvard Medical School, Boston, MA

First Author:

Michael Ferguson  
Harvard Medical School
Boston, MA

Co-Author(s):

Vicky Chen  
Massachusetts Institute of Technology
Cambridge, MA
Joseph Turner  
Brigham and Women's Hospital
Boston, MA
Anish Suvarna  
Thomas Jefferson High School for Science and Technology
Alexandria, VA
Eli Baughan  
Brigham Young University
Provo, UT
Kiana Bunnell  
Brigham Young University
Provo, UT
Jared Nielsen  
Brigham Young University
Provo, UT
Frederic Schaper  
Brigham and Women’s Hospital, Harvard Medical School
Boston, MA

Introduction:

Given the recent surge of methodological innovation using human brain lesion data, our objective is to create an open source, web-based platform for aggregating, viewing, and analyzing published case reports containing both brain imaging and clinical evaluation of the patient.

Methods:

LesionBank.org offers a user-friendly interface for exploring a collection of curated brain lesion case reports. LesionBank enables users to search, visualize, and retrieve lesion images and related metadata. The application leverages the Django web-framework and a Postgres database to process user requests and handle the user interface. The application is deployed on a DigitalOcean droplet, and imaging data is stored in an S3-compatible DigitalOcean object space. Currently, the collection includes 163 lesion ROIs and associated lesion network maps that were found from previous published case reports.

Results:

LesionBank.org has been launched with a viewer and search capabilities based on both textual search of case reports and image-based search of published brain lesion images.To date, the LesionBank.org platform has been used to successfully reproduce primary findings from an already-published study on amnesia (Ferguson, 2019). Additionally, LesionBank.org has been used to support training of several dozen undergraduate research assistants to identify brain lesion case reports of interest, create digital lesion tracings from published, catalog metadata, and relate brain lesions to their underlying functional connectivity.

Conclusions:

Science: LesionBank, an open source platform for brain lesion case reports, is able to reproduce brain lesion mapping results from published literature.
Education: LesionBank is able to power an asynchronous undergraduate semester research course on clinical neuroscience.

Education, History and Social Aspects of Brain Imaging:

Education, History and Social Aspects of Brain Imaging

Neuroinformatics and Data Sharing:

Databasing and Data Sharing 1
Informatics Other 2

Keywords:

Computing
DISORDERS
Open Data
Other

1|2Indicates the priority used for review

Provide references using author date format

Fox, M. D. (2018). 'Mapping symptoms to brain networks with the human connectome', New England Journal of Medicine, vol. 379, no. 23, pp. 2237-2245.