Segmentation method for cerebral blood vessels from MRA using hysteresis
This work addresses the need for open-source tools to generate training data for deep learning models in medical imaging, specifically for cerebral blood vessel segmentation across modalities, though it is incremental as it builds on classical techniques.
The authors tackled the problem of scarce annotated data for cerebral blood vessel segmentation from MRI by developing a classical segmentation method using hysteresis thresholding to generate ground truth for deep learning training. Their method achieved a high vessel segmentation quality score of 14.2 out of 15 in clinician evaluations on 24 3D images.
Segmentation of cerebral blood vessels from Magnetic Resonance Imaging (MRI) is an open problem that could be solved with deep learning (DL). However, annotated data for training is often scarce. Due to the absence of open-source tools, we aim to develop a classical segmentation method that generates vessel ground truth from Magnetic Resonance Angiography for DL training of segmentation across a variety of modalities. The method combines size-specific Hessian filters, hysteresis thresholding and connected component correction. The optimal choice of processing steps was evaluated with a blinded scoring by a clinician using 24 3D images. The results show that all method steps are necessary to produce the highest (14.2/15) vessel segmentation quality score. Omitting the connected component correction caused the largest quality loss. The method, which is available on GitHub, can be used to train DL models for vessel segmentation.