Assessment of Takayasu's arteritis activity by ultrasound localization microscopy

Figures

• 

  Central Illustration. Experimental design of the study and analysis of the ultrasound localization microscopy data in comparison with the NIH Takayasu arteritis activity score. NIH: National Institute of Health. A boxplot comparing the number of microbubbles tracked in the carotid wall is shown in the lower left. It shows a clear separation between patients with active disease from those with quiescent disease.

• 

  Fig. 1. Summary diagram of the processing method. (A) Acquisition: ultrafast ultrasound imaging acquisition coupled with intravenous injection of microbubbles, raw radio frequency (RF) data are saved and processed offline. (B) Beamforming: ultrasound In-phase Quadrature (IQ) image reconstruction with delay and sum beamforming. (C) Data processing: 1. pulsatility detection based on carotid wall motion; 2. selection of blocks of 100 frames with less pulsatility; 3. microbubbles filtering from tissue signal based on Singular Value Decomposition (SVD); 4. Power Doppler to retrieve the lumen in each block; 5. motion correction with lumen registration between blocks. (D) Ultrasound localization microscopy on each block of filtered frames: 1. localisation of the microbubbles; 2. tracking of their positions. E. Registration of the microbubbles tracks' positions from the blocks' lumen and accumulation of the tracks over the whole acquisition.

• 

  Fig. 2. Quantification of the microbubbles (MB) within the carotid wall using ultrasound localization microscopy (ULM) in patients with active or quiescent Takayasu arteritis: number of MB tracked (A), sum of the length of the tracks (L score) (B) and perfusion score (C). P-values are obtained using the Mann–Whitney test.

• 

  Fig. 3. Graph of microbubble velocity distribution within the arterial wall of patients with active Takayasu arteritis (n = 5).

• 

  Fig. 4. Example of a case of active Takayasu arteritis with CTA reconstruction in frontal view of the right common carotid artery (A). In B is adapted the image at the same site obtained in ultrasound localization microscopy (obtained from 8 s long acquisition after motion correction) with visualisation of microvessels (B). FDG-PET/CT imaging shows intense fixation localised to the common carotid walls (C).

• 

  Fig. 5. Ultrasound localization microscopy of the common carotid wall of the same patient with initially quiescent Takayasu arteritis (A), and with a reactivation of the disease 9 months later with new imaging showing a significant passage of microbubbles in the arterial wall (B), from 8 s long acquisition after motion correction and Ultrasound localization microscopy data accumulated.