TY - JOUR
T1 - Evolution of white matter hyperintensity segmentation methods and implementation over the past two decades; an incomplete shift towards deep learning
AU - CSF and Blood
AU - Genetics
AU - Neuropathology
AU - NACC
AU - DIAN
AU - Cerebrovascular Disease (CVD) Risk
AU - Hippocampal Sclerosis (HS-TDP43) Risk
AU - Artificial Intelligence and Machine Learning
AU - for the ADOPIC, ADNI Investigators
AU - Biostats, Database and Bioinformatics
AU - Cognition
AU - Imaging
AU - Rahmani, Maryam
AU - Dierker, Donna
AU - Yaeger, Lauren
AU - Saykin, Andrew
AU - Luckett, Patrick H.
AU - Vlassenko, Andrei G.
AU - Owens, Christopher
AU - Jafri, Hussain
AU - Womack, Kyle
AU - Fripp, Jurgen
AU - Xia, Ying
AU - Tosun, Duygu
AU - Benzinger, Tammie L.S.
AU - Masters, Colin L.
AU - Lee, Jin Moo
AU - Morris, John C.
AU - Goyal, Manu S.
AU - Strain, Jeremy F.
AU - Kukull, Walter
AU - Weiner, Michael
AU - Burnham, Samantha
AU - CoxDoecke, Tim James
AU - Fedyashov, Victor
AU - Shishegar, Rosita
AU - Xiong, Chengjie
AU - Marcus, Daniel
AU - Raniga, Parnesh
AU - Li, Shenpeng
AU - Aschenbrenner, Andrew
AU - Hassenstab, Jason
AU - Lim, Yen Ying
AU - Maruff, Paul
AU - Sohrabi, Hamid
AU - Robertson, Jo
AU - Markovic, Shaun
AU - Bourgeat, Pierrick
AU - Doré, Vincent
AU - Mayo, Clifford Jack
AU - Mussoumzadeh, Parinaz
AU - Rowe, Chris
AU - Villemagne, Victor
AU - Bateman, Randy
AU - Fowler, Chris
AU - Li, Qiao Xin
AU - Martins, Ralph
AU - Schindler, Suzanne
AU - Shaw, Les
AU - Cruchaga, Carlos
AU - Harari, Oscar
AU - Laws, Simon
AU - Porter, Tenielle
AU - O’Brien, Eleanor
AU - Perrin, Richard
AU - McDade, Eric
AU - Jack, Clifford
AU - Morris, John
AU - Yassi, Nawaf
AU - Roberts, Blaine
AU - Goudey, Benjamin
N1 - Publisher Copyright:
© The Author(s) 2024.
PY - 2024/10
Y1 - 2024/10
N2 - This systematic review examines the prevalence, underlying mechanisms, cohort characteristics, evaluation criteria, and cohort types in white matter hyperintensity (WMH) pipeline and implementation literature spanning the last two decades. Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, we categorized WMH segmentation tools based on their methodologies from January 1, 2000, to November 18, 2022. Inclusion criteria involved articles using openly available techniques with detailed descriptions, focusing on WMH as a primary outcome. Our analysis identified 1007 visual rating scales, 118 pipeline development articles, and 509 implementation articles. These studies predominantly explored aging, dementia, psychiatric disorders, and small vessel disease, with aging and dementia being the most prevalent cohorts. Deep learning emerged as the most frequently developed segmentation technique, indicative of a heightened scrutiny in new technique development over the past two decades. We illustrate observed patterns and discrepancies between published and implemented WMH techniques. Despite increasingly sophisticated quantitative segmentation options, visual rating scales persist, with the SPM technique being the most utilized among quantitative methods and potentially serving as a reference standard for newer techniques. Our findings highlight the need for future standards in WMH segmentation, and we provide recommendations based on these observations.
AB - This systematic review examines the prevalence, underlying mechanisms, cohort characteristics, evaluation criteria, and cohort types in white matter hyperintensity (WMH) pipeline and implementation literature spanning the last two decades. Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, we categorized WMH segmentation tools based on their methodologies from January 1, 2000, to November 18, 2022. Inclusion criteria involved articles using openly available techniques with detailed descriptions, focusing on WMH as a primary outcome. Our analysis identified 1007 visual rating scales, 118 pipeline development articles, and 509 implementation articles. These studies predominantly explored aging, dementia, psychiatric disorders, and small vessel disease, with aging and dementia being the most prevalent cohorts. Deep learning emerged as the most frequently developed segmentation technique, indicative of a heightened scrutiny in new technique development over the past two decades. We illustrate observed patterns and discrepancies between published and implemented WMH techniques. Despite increasingly sophisticated quantitative segmentation options, visual rating scales persist, with the SPM technique being the most utilized among quantitative methods and potentially serving as a reference standard for newer techniques. Our findings highlight the need for future standards in WMH segmentation, and we provide recommendations based on these observations.
KW - Deep Learning
KW - Pipelines
KW - Segmentation
KW - Systematic Review
KW - White matter hyperintensity
UR - http://www.scopus.com/inward/record.url?scp=85204204426&partnerID=8YFLogxK
U2 - 10.1007/s11682-024-00902-w
DO - 10.1007/s11682-024-00902-w
M3 - Review article
C2 - 39083144
AN - SCOPUS:85204204426
SN - 1931-7557
VL - 18
SP - 1310
EP - 1322
JO - Brain Imaging and Behavior
JF - Brain Imaging and Behavior
IS - 5
ER -