For SHARP+MEDI pipelines, median percentage difference in effect sizes from (a) different reference regions with FMRIB’s Integrated Registration and Segmentation Tool (FIRST) segmentation, relative to the originally used frontal deep white matter (FDWM) reference region, and (b) the Bi-parametric Joint Label Fusion (B-JLF) segmentation method, both across all reference regions (All) and individually for each reference region (FDWM, whole brain [WB], and cerebrospinal fluid [CSF]), relative to FIRST and their respective counterparts.

For SHARP+MEDI pipelines, median percentage difference in effect sizes from (a) different reference regions with FMRIB’s Integrated Registration and Segmentation Tool (FIRST) segmentation, relative to the originally used frontal deep white matter (FDWM) reference region, and (b) the Bi-parametric Joint Label Fusion (B-JLF) segmentation method, both across all reference regions (All) and individually for each reference region (FDWM, whole brain [WB], and cerebrospinal fluid [CSF]), relative to FIRST and their respective counterparts.

<p>Median percentage differences are indicated outside the bars, with negative values indicating a decrease in absolute effect sizes and positive values indicating an increase.</p>

Saved in:

Bibliographic Details
Main Author:	Fahad Salman (6150395) (author)
Other Authors:	Niels Bergsland (105411) (author), Michael G. Dwyer (9134261) (author), Jack A. Reeves (8292495) (author), Abhisri Ramesh (16990329) (author), Dejan Jakimovski (5515748) (author), Bianca Weinstock-Guttman (105396) (author), Robert Zivadinov (105393) (author), Ferdinand Schweser (392834) (author)
Published:	2025
Subjects:	Biochemistry Medicine Cell Biology Genetics Biotechnology Evolutionary Biology Developmental Biology Cancer Infectious Diseases Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified yielded inconsistent results phenomenological &# 8220 explain contradicting findings dipole inversion algorithms decline &# 8221 three reference regions 44 healthy controls recent replication study previous replication study xlink "> using two segmentation methods thalamic susceptibility assessment pwms previously reported different qsm algorithms xlink "> thalamic susceptibility two qsm segmentation method reference region controls independent unique combination tested combinations specific parameter rise late regularization parameters pwms compared multiple sclerosis mri dataset distinct pipeline chosen methodology based studies authors failed
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Cannot write session to /tmp/vufind_sessions/sess_4iu3a7rujlg63ev1j5qc4829o2