Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Topology-Aware Exploration of Circle of Willis for CTA and MRA: Segmentation, Detection, and Classification

Minghui Zhang, Xin You, Hanxiao Zhang, Yun Gu

TL;DR

This work utilizes the topology-aware loss to enhance the topology completeness of the CoW and the discrimination between different classes, and a complementary topology-aware refinement is further conducted to enhance the connectivity within the same class.

Abstract

The Circle of Willis (CoW) vessels is critical to connecting major circulations of the brain. The topology of the vascular structure is clinical significance to evaluate the risk, severity of the neuro-vascular diseases. The CoW has two representative angiographic imaging modalities, computed tomography angiography (CTA) and magnetic resonance angiography (MRA). TopCow24 provided 125 paired CTA-MRA dataset for the analysis of CoW. To explore both CTA and MRA images in a unified framework to learn the inherent topology of Cow, we construct the universal dataset via independent intensity preprocess, followed by joint resampling and normarlization. Then, we utilize the topology-aware loss to enhance the topology completeness of the CoW and the discrimination between different classes. A complementary topology-aware refinement is further conducted to enhance the connectivity within the same class. Our method was evaluated on all the three tasks and two modalities, achieving competitive results. In the final test phase of TopCow24 Challenge, we achieved the second place in the CTA-Seg-Task, the third palce in the CTA-Box-Task, the first place in the CTA-Edg-Task, the second place in the MRA-Seg-Task, the third palce in the MRA-Box-Task, the second place in the MRA-Edg-Task.

Topology-Aware Exploration of Circle of Willis for CTA and MRA: Segmentation, Detection, and Classification

TL;DR

This work utilizes the topology-aware loss to enhance the topology completeness of the CoW and the discrimination between different classes, and a complementary topology-aware refinement is further conducted to enhance the connectivity within the same class.

Abstract

The Circle of Willis (CoW) vessels is critical to connecting major circulations of the brain. The topology of the vascular structure is clinical significance to evaluate the risk, severity of the neuro-vascular diseases. The CoW has two representative angiographic imaging modalities, computed tomography angiography (CTA) and magnetic resonance angiography (MRA). TopCow24 provided 125 paired CTA-MRA dataset for the analysis of CoW. To explore both CTA and MRA images in a unified framework to learn the inherent topology of Cow, we construct the universal dataset via independent intensity preprocess, followed by joint resampling and normarlization. Then, we utilize the topology-aware loss to enhance the topology completeness of the CoW and the discrimination between different classes. A complementary topology-aware refinement is further conducted to enhance the connectivity within the same class. Our method was evaluated on all the three tasks and two modalities, achieving competitive results. In the final test phase of TopCow24 Challenge, we achieved the second place in the CTA-Seg-Task, the third palce in the CTA-Box-Task, the first place in the CTA-Edg-Task, the second place in the MRA-Seg-Task, the third palce in the MRA-Box-Task, the second place in the MRA-Edg-Task.

Paper Structure

This paper contains 11 sections, 3 equations, 3 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Method
  3. Universal Dataset Construction
  4. Topology-aware Loss
  5. Topology-aware Refinement
  6. Experimental Analysis
  7. Implementation Detail
  8. Task-1: Multi-class Segmentation of CoW (CTA and MRI)
  9. Task-2: CoW RoI Detection (CTA and MRI)
  10. Task-3: CoW graph Classification (CTA and MRI)
  11. Discussion and Future Directions

Figures (3)

  • Figure 1: The process to acquire the universal dataset from the CTA and MRI scans. $[-1000,1800]$ is used to truncate the CTA scans. As for the MRI scans, we chose $[0,700]$. After independently processing the intensity truncation, the modality-agnostic scans were constructed. These modality-agnostic scans are then universally resampled to the same resolution. Finally, we normalize the intensity of each case to $[0, 1]$.
  • Figure 2: The visualization of the dynamic weights of the connectivity-aware loss for multi-class vessels. The weighting is conducted on-the-fly based on the patch volume and independently calculated for each class.
  • Figure 3: The qualitative results of the topology-aware refinement.