Deep-learning based tractography for neonates

Sovanlal Mukherjee, Natacha Paquette, Niharika Gajawelli, Yalin Wang, Julia Wallace, Marvin D. Nelson, Ashok Panigrahy, Natasha Lepore

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Generation of white matter (WM) tractography for neonates has primarily depended on a successful development of a diffusion tensor imaging (DTI)-based ATLAS. In this study, we present a deep-learning framework for WM tractography of neonates’ brains that is independent of any specific ATLAS. A convolutional neural network (CNN)-based deep-learning architecture is proposed for automated generation of WM tractography. Our dataset consists of DWI scans of 40 neonates that were used to train the model. Although the proposed model is adopted for WM tractography, it can generally be applied for subcortical structures and cerebellum.

Original languageEnglish (US)
Title of host publication16th International Symposium on Medical Information Processing and Analysis
EditorsEduardo Romero, Natasha Lepore, Jorge Brieva, Marius Linguraru
PublisherSPIE
ISBN (Electronic)9781510639911
DOIs
StatePublished - 2020
Event16th International Symposium on Medical Information Processing and Analysis 2020 - Lima, Virtual, Peru
Duration: Oct 3 2020Oct 4 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11583
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference16th International Symposium on Medical Information Processing and Analysis 2020
Country/TerritoryPeru
CityLima, Virtual
Period10/3/2010/4/20

Keywords

  • Convolutional neural network
  • Deep-learning
  • Tractography
  • White-matter

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Deep-learning based tractography for neonates'. Together they form a unique fingerprint.

Cite this