PreOperative Performance

Sharing work presented at the @esmrmb.bsky.social meeting in Marseille earlier this month, demonstrating our anisotropic diffusion phantom’s utility with higher-order tensor diffusion data models.

#Neuroscience #neuro #diffusionMRI #mri #dki #ESMRMB2025

PreOperative Performance is delighted to attend the International Society of Tractography meeting in Bordeaux this week!

Say hello, DM or email

✉️: contact@preoperativeperformance.com

@tractography.bsky.social

Some last scenes from near Marseille’s Vieux Port. Marseille was founded ~ 600BCE by Phoenician sailors and despite modern infrastructure requirements has wonderfully preserved layers of history.

Thank you @esmrmb.bsky.social and thank you Marseille!

This was a greatly enjoyable meeting and I was delighted to see sessions dedicated to the importance of medical imaging quality (control & assurance) and reference standards.

(1/2)

Poster day! @esmrmb.bsky.social

#MRI #diffusionMRI #DKI #HARDI #neuroscience #neurorad #neurosky

(3/3)

As these imaging modalities are sensitive to noise and inter system variability, reference standards such as our anisotropic diffusion phantom can verify repeatability and consistency of measurements, with applications in #biomarker characterisation, and multi-system and longitudinal studies.

(2/3)

Day 3 of #ESMRMB2025 begins shortly. It’s poster day!

In poster PG353 PreOperative Performance’s technology is applied to higher order tensor models of diffusion MRI data which can resolve complex neural architecture such as areas with crossing fibers.

(1/3)

#ESMRMB2025 #microstructuralimaging #UHF #MRI #radsky #diffusionMRI

Join us at the preclinical imaging workshop session on Wednesday 8th at #ESMRMB meeting in Marseille.

(4/4)

PreOperative Performance has metrology phantoms, compatible with NMR and ultra-high field (UHF) imaging workflows, that can support this validation, provide quality controls that inform human-studies, and strengthen interpretability and impact.

(3/4)

…that can be incorporated into human-centered investigations.

The characterisation of imaging features, structures and biomarkers, relevant to understanding disease pathways and progression, require validation solutions.

(2/4)

Preclinical imaging, including studies of ex-vivo tissue and organoid models, play a critical role in advancing MR imaging research by providing the means to explore the biological underpinnings of disease, refine imaging techniques and uncover insights…

🧵1/n

Ensure you take some time to check out abstract "Assessing measurement consistency of a novel anisotropic phantom for higher order diffusion tensor MRI sequences' (PG353).

If you’re attending, we’d love to meet in Marseille!

Follow us & DM to connect!

(2/2)

We’re looking forward to attending #ESMRMB 2025 from October 8-11th
🧠

We look forward to meeting old friends and new and engaging in discussions about our anisotropic diffusion phantom technology. We are delighted to share that our technology will feature in the scientific program.

1/2

This image shows 3 pairs of crossing white matter simulating modules with 3 different bundle diameters from our phantom POP-0005. Note that the bundles with smallest diameter are sub-voxel dimensions.

🧵(1/3)

MR imaging at 1.5T vs 3T have different advantages. 🧠

Aside from cost considerations, imaging at 3T delivers higher signal to noise (SNR) enabling increased spatial resolution, critical for identification of smaller white matter tracts and changes that may result from injury or disease.

1/8

Through use of PreOperative Performance’s phantom technology, we can explain tractography performance and determine the resolvability of tracts that can be achieved given their sizes, specific to the MR system and protocol used.

8/8

In contrast, at 1.5T, several seeds were required on most bundles and the thinnest bundles were not trackable.

7/8

The 3T data quality and protocol resulted in well-defined tracts (all 6 with 3 varying diameters ) when seed-based probabilistic tractography was performed. Multiple seeds were only required for the thinner bundles.

6/8

In particular, for 1.5T data, there is a lot of green in the horizontal bundles on slice #52 which is why it was hard to obtain tracking for those as shown in B (3 seeds needed: left, middle and right) to get to termination points (yellow spheres).

5/8

Data were collected with same resolution and b-value (b=1000 s/mm^2) but different number of gradient directions. We can see that the 3T V1 results are much less noisy (all 3 slices show exclusively green and red in the vertical and horizontal bundles, respectively).

4/8

In this diagram we show a comparison of results for fitting the tensor model to data collected at 2 field strengths and the resulting tractography.

3/8

Imaging at 1.5T is not without its advantages, there are less distortions at lower field and some patients with implanted medical devices (such as cochlear implants) may be conditionally scanned at 1.5T and not at higher field strengths.

2/8

MR imaging at 1.5T vs 3T have different advantages. 🧠

Aside from cost considerations, imaging at 3T delivers higher signal to noise (SNR) enabling increased spatial resolution, critical for identification of smaller white matter tracts and changes that may result from injury or disease.

1/8

🫶

@ohbmofficial.bsky.social

A post on visualizing white matter bifurcations using diffusion analysis:
🧠

We show that using the diffusion model and seed-based #tractography, we are able to delineate the full bifurcating module of a bifurcating module in our phantom POP-0005 with dimensions as given.

🧵(1/4)

#mri #dmri #medicalimaging #neurorad #neuroimaging #neurosky

The scan was performed on a 3T
#MAGNUS system by GE Healthcare with high in-plane isotropic resolution (0.937 mm) and with 2.0mm slices, 3 non-diffusion weighted volumes and 50 gradient directions (b-value = 1000 s/mm2 ).

(4/4)

We used FSL's bedpost and Fdt with spherical mask seed points (shown in yellow) and termination sphere masks shown in red. The 3D rendering shows the result as a mask.

(3/4)