Open AccessTranscranial Doppler-Based Surrogates for Cerebral Blood Flow: A Statistical Study
Author(s) -
Joseph Hart,
Vera Novak,
Charles Saunders,
Pierre A. Gremaud
Publication year - 2016
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0165536
Subject(s) - transcranial doppler , cerebral blood flow , magnetic resonance imaging , blood flow , medicine , estimator , cardiology , middle cerebral artery , nuclear magnetic resonance , radiology , physics , mathematics , statistics , ischemia
It is commonly assumed that perfusion in a given cerebral territory can be inferred from Blood Flow Velocity (BFV) measurements in the corresponding stem artery. In order to test this hypothesis, we construct a cerebral blood flow (CBF) estimator based on transcranial Doppler (TCD) blood flow velocity and ten other easily available patient characteristics and clinical parameters. A total of 261 measurements were collected from 88 older patients. The estimator is based on local regression (Random Forest). Its performance is analyzed against baseline CBF from 3-D pseudocontinuous arterial spin labeling (pCASL) magnetic resonance imaging (MRI). Patient specific CBF predictions are of poor quality ( r = 0.41 and p -value = 4.5 × 10 −12 ); the hypothesis is thus not clearly supported by evidence.