Related Experiment Video
Updated: May 15, 2026

Assessing Cortical Cerebral Microinfarcts on High Resolution MR Images
Published on: November 20, 2015
Detection of Hemorrhagic Transformation of Cerebral Infarcts with Ultra-low Field MRI
Dimah Hasan1, Julian Sauer1, Clara Heller1
1Department of Neuroradiology, Universitätsklinikum Aachen, Aachen, Germany.
Purpose:
Hemorrhagic transformation (HT) within ischemic infarcts is a critical post-stroke complication that significantly influences secondary prevention strategies, particularly the timing and safety of anticoagulation. Ultra-low field (ULF) MRI (0.064 T, Swoop, Hyperfine) offers bedside neuroimaging, but its ability to detect hemorrhagic transformation compared to standard MRI (1.5 T or 3 T) and CT remains uncertain. This study aimed to evaluate the diagnostic accuracy of ULF-pMRI for detecting hemorrhagic transformation in ischemic stroke patients.
Methods:
We retrospectively analyzed 97 patients with ischemic infarcts who underwent both ULF-pMRI and standard neuroimaging (standard MRI or CT). The presence of hemorrhagic transformation was assessed by two independent readers (trained neuroradiologists) and by consensus reading in univariate analysis using the Heidelberg Bleeding Classification (HBC). Diagnostic accuracy metrics sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated with the standard method (standard MRI or CT within two days of ULF-pMRI) as the reference. Agreement was quantified using Cohen's Kappa.
Results:
Of 97 stroke patients, n = 21 (21.6%) showed hemorrhagic transformation on standard imaging. Most findings corresponded to scattered small petechia and subarachnoid hemorrhage. Reader 1 detected hemorrhagic transformation in ULF-pMRI with a sensitivity of 5/21 (23.8%), specificity 100%, positive predictive value (PPV) 100%, and negative predictive value (NPV) 82% and Reader 2 showed a sensitivity of 3/21 (14.3%), specificity 100%, PPV 100%, and NPV 81%, while consensus reading showed sensitivity of 3/21 (14.3%), specificity 100%, PPV 100%, NPV 81%. ULF-pMRI failed to detect petechial hemorrhages (0/13 HBC1a; 0/3 HBC1b), while detection improved for more extensive hemorrhagic transformations (HBC 1c and 3). After exclusion of petechial hemorrhages (HBC1a and HBC1b), ULF-pMRI detected 5 of 9 remaining hemorrhagic transformation subtypes (55.6%).
Conclusion:
Ultra-low-field MRI demonstrated low sensitivity for detecting hemorrhagic transformation in ischemic stroke, particularly for petechial hemorrhages (HBC class 1a and 1b). Detection was higher for non-petechial hemorrhagic transformations (HBC class 1c and 3), but overall performance remains insufficient for reliable therapy-guiding decisions.
More Related Videos
05:23Construction and Application of Cerebral Functional Region-Based Cerebral Blood Flow Atlas Using Magnetic Resonance Imaging-Arterial Spin Labeling
Published on: May 31, 2024
09:06Whole-brain Segmentation and Change-point Analysis of Anatomical Brain MRI—Application in Premanifest Huntington's Disease
Published on: June 9, 2018