Related Experiment Video
Updated: Jun 4, 2026
08:12
A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage
Published on: July 28, 2018
Classification-based summation of cerebral digital subtraction angiography series for image post-processing
D Schuldhaus1, M Spiegel, T Redel
1University Erlangen-Nuremberg, Department of Neuroradiology, Erlangen, Germany.
Physics in Medicine and Biology
|February 25, 2011
Summary
This study introduces an automated method to process digital subtraction angiography (DSA) series for cerebrovascular disease imaging. The technique enhances diagnostic clarity by creating a single, composite image from multiple DSA phases, improving vessel visualization and reducing artifacts.
Area of Science:
- Medical Imaging
- Neuroradiology
- Image Processing
Background:
- 2D digital subtraction angiography (DSA) is crucial for diagnosing and planning cerebrovascular disease treatments.
- Current DSA series often lack a single image capturing the entire vessel tree, complicating post-processing for algorithms.
- Image processing tasks like segmentation and registration are hindered by the fragmented nature of DSA data.
Purpose of the Study:
- To develop a novel method for automatically splitting digital subtraction angiography (DSA) series into distinct phases (mask, arterial, parenchymal).
- To generate a single, composite image from the arterial phase of DSA series, enhancing visualization of the entire cerebrovascular network.
- To reduce noise and motion artifacts in DSA images for improved downstream analysis.
Main Methods:
- A two-step automated phase classification approach was employed.
- A Perceptron-based method determined the mask/arterial phase boundary.
- A threshold-based method identified the arterial/parenchymal phase boundary.
- Image summation or minimum intensity projection was used to create the final composite image.
Main Results:
- The automated method achieved a 93% match with expert phase separation for the mask/arterial border and 50% for the arterial/parenchymal border.
- The final composite image demonstrated a significant increase in signal-to-noise ratio (SNR) by up to 182% compared to summing the entire series.
- The method effectively combined arterial phase images, reducing noise and motion artifacts.
Conclusions:
- The proposed automated method accurately separates DSA phases, facilitating the creation of enhanced composite images.
- This technique significantly improves image quality, offering a more comprehensive view of the cerebrovascular system for diagnosis and treatment planning.
- The automated approach simplifies complex image processing tasks, potentially reducing the burden on medical experts.
