OpenLB
Open Library of Bioscience
Advanced Search
Magnetic resonance imaging
06, 2019;59 143-152.

Registration-based image enhancement improves multi-atlas segmentation of the thalamic nuclei and hippocampal subfields.

Shunxing Bao, Camilo Bermudez, Yuankai Huo, Prasanna Parvathaneni, William Rodriguez, Susan M Resnick, Pierre-François D'Haese, Maureen McHugo, Stephan Heckers, Benoit M Dawant, Ilwoo Lyu, Bennett A Landman
Author Information
  1. Shunxing Bao: Computer Science, Vanderbilt University, Nashville, TN, United States of America. Electronic address: shunxing.bao@vanderbilt.edu.
  2. Camilo Bermudez: Biomedical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  3. Yuankai Huo: Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  4. Prasanna Parvathaneni: Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  5. William Rodriguez: Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  6. Susan M Resnick: Laboratory of Behavioral Neuroscience, National Institute on Aging, MD, United States of America.
  7. Pierre-François D'Haese: Computer Science, Vanderbilt University, Nashville, TN, United States of America; Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America; Department of Neurological Surgery, Vanderbilt University Medical Center, Nashville, TN, United States of America.
  8. Maureen McHugo: Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN, United States of America.
  9. Stephan Heckers: Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN, United States of America.
  10. Benoit M Dawant: Computer Science, Vanderbilt University, Nashville, TN, United States of America; Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America; Biomedical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  11. Ilwoo Lyu: Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America.
  12. Bennett A Landman: Computer Science, Vanderbilt University, Nashville, TN, United States of America; Electrical Engineering, Vanderbilt University, Nashville, TN, United States of America; Biomedical Engineering, Vanderbilt University, Nashville, TN, United States of America; Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN, United States of America.
PMID: 30880111 DOI: 10.1016/j.mri.2019.03.014

Abstract

Magnetic resonance imaging (MRI) is an important tool for analysis of deep brain grey matter structures. However, analysis of these structures is limited due to low intensity contrast typically found in whole brain imaging protocols. Herein, we propose a big data registration-enhancement (BDRE) technique to augment the contrast of deep brain structures using an efficient large-scale non-rigid registration strategy. Direct validation is problematic given a lack of ground truth data. Rather, we validate the usefulness and impact of BDRE for multi-atlas (MA) segmentation on two sets of structures of clinical interest: the thalamic nuclei and hippocampal subfields. The experimental design compares algorithms using T1-weighted 3 T MRI for both structures (and additional 7 T MRI for the thalamic nuclei) with an algorithm using BDRE. As baseline comparisons, a recent denoising (DN) technique and a super-resolution (SR) method are used to preprocess the original 3 T MRI. The performance of each MA segmentation is evaluated by the Dice similarity coefficient (DSC). BDRE significantly improves mean segmentation accuracy over all methods tested for both thalamic nuclei (3 T imaging: 9.1%; 7 T imaging: 15.6%; DN: 6.9%; SR: 16.2%) and hippocampal subfields (3 T T1 only: 8.7%; DN: 8.4%; SR: 8.6%). We also present DSC performance for each thalamic nucleus and hippocampal subfield and show that BDRE can help MA segmentation for individual thalamic nuclei and hippocampal subfields. This work will enable large-scale analysis of clinically relevant deep brain structures from commonly acquired T1 images.

Keywords

Big data Deep brain structure Non-rigid registration

References

  1. Nat Rev Neurosci. 2007 Aug;8(8):623-35 [PMID: 17637800]
  2. J Neurosurg. 2010 Sep;113(3):639-47 [PMID: 20380532]
  3. Med Image Anal. 2010 Aug;14(4):594-605 [PMID: 20580893]
  4. Magn Reson Imaging. 2010 Dec;28(10):1485-96 [PMID: 20850239]
  5. Med Image Comput Comput Assist Interv. 2010;13(Pt 2):355-62 [PMID: 20879335]
  6. Nat Rev Neurosci. 2011 Sep 07;12(10):585-601 [PMID: 21897434]
  7. IEEE Trans Pattern Anal Mach Intell. 2013 Mar;35(3):611-23 [PMID: 22732662]
  8. Comput Methods Programs Biomed. 2013 Jan;109(1):65-73 [PMID: 23036854]
  9. Med Image Comput Comput Assist Interv. 2013;16(Pt 1):187-94 [PMID: 24505665]
  10. Hum Brain Mapp. 2015 Jan;36(1):258-87 [PMID: 25181316]
  11. Neurosci Biobehav Rev. 2015 Jul;54:57-75 [PMID: 25616183]
  12. Med Image Anal. 2015 May;22(1):35-47 [PMID: 25725303]
  13. Neuroimage. 2015 Jul 15;115:117-37 [PMID: 25936807]
  14. Neuroimage. 2016 Jan 1;124(Pt B):1097-1101 [PMID: 25988229]
  15. Med Image Anal. 2016 Jan;27:93-104 [PMID: 26096982]
  16. IEEE Trans Med Imaging. 2016 Sep;35(9):2085-97 [PMID: 27046894]
  17. Med Image Comput Comput Assist Interv. 2016 Oct;9900:81-88 [PMID: 28191550]
  18. IEEE Trans Med Imaging. 2018 Feb;37(2):384-395 [PMID: 28961105]
  19. IEEE Trans Med Imaging. 2018 Jun;37(6):1297-1309 [PMID: 29870360]
  20. Proc SPIE Int Soc Opt Eng. 2018 Mar;10597:null [PMID: 29887668]

Grants

  1. UL1 TR000445/NCATS NIH HHS
  2. P30 AG066507/NIA NIH HHS
  3. R21 EY024036/NEI NIH HHS
  4. ZIA AG000191-22/Intramural NIH HHS
  5. UL1 RR024975/NCRR NIH HHS
  6. R01 NS095291/NINDS NIH HHS
  7. ZIA AG000191-21/Intramural NIH HHS
  8. R21 NS064534/NINDS NIH HHS

MeSH Term

Algorithms
Brain Mapping
Hippocampus
Humans
Image Enhancement
Image Processing, Computer-Assisted
Magnetic Resonance Imaging
Temporal Lobe
Thalamic Nuclei
Journal Article Research Support, N.I.H., Extramural Research Support, N.I.H., Intramural Research Support, U.S. Gov't, Non-P.H.S.
Article has an altmetric score of 1

Word Cloud

Created with Highcharts 10.0.0structuresthalamicbrainBDREsegmentationnucleihippocampalMRIsubfields3 TanalysisdeepdatausingMA8imagingcontrasttechniquelarge-scaleregistrationmulti-atlas7 TperformanceDSCimprovesimaging:6%DN:SR:T1MagneticresonanceimportanttoolgreymatterHoweverlimitedduelowintensitytypicallyfoundwholeprotocolsHereinproposebigregistration-enhancementaugmentefficientnon-rigidstrategyDirectvalidationproblematicgivenlackgroundtruthRathervalidateusefulnessimpacttwosetsclinicalinterest:experimentaldesigncomparesalgorithmsT1-weightedadditionalalgorithmbaselinecomparisonsrecentdenoisingDNsuper-resolutionSRmethodusedpreprocessoriginalevaluatedDicesimilaritycoefficientsignificantlymeanaccuracymethodstested91%1569%162%only:7%4%alsopresentnucleussubfieldshowcanhelpindividualworkwillenableclinicallyrelevantcommonlyacquiredimagesRegistration-basedimageenhancementBigDeepstructureNon-rigid

Similar Articles

Cited By (5)