Low-Field Combined Diffusion-Relaxation MRI for Mapping Placenta Structure and Function

Abstract

Purpose: Demonstrating quantitative multi-parametric mapping in the placenta with combined $T{2}^{\ast }$-diffusion MRI at low-field (0.55T).

Methods: We present 57 placental MRI scans performed on a commercially available 0.55T scanner. We acquired the images using a combined $T{2}^{\ast }$-diffusion technique scan that simultaneously acquires multiple diffusion preparations and echo times. We processed the data to produce quantitative $T{2}^{\ast }$ and diffusivity maps using a combined $T{2}^{\ast }$-ADC model. We compared the derived quantitative parameters across gestation in healthy controls and a cohort of clinical cases.

Results: Quantitative parameter maps closely resemble those from previous experiments at higher field strength, with similar trends in $T{2}^{\ast }$ and ADC against gestational age observed.

Conclusion: Combined $T{2}^{\ast }$-diffusion placental MRI is reliably achievable at 0.55T. The advantages of lower field strength - such as cost, ease of deployment, increased accessibility and patient comfort due to the wider bore, and increased $T{2}^{\ast }$ for larger dynamic ranges - can support the widespread roll out of placental MRI as an adjunct to ultrasound during pregnancy.

Figure 1:

Exemplary direction-averaged data for three TEs and four b-values after denoising for a single control participant with GA = 27.14 weeks. See Figure 2 for derived maps for this participant.

Figure 2:

Detailed view into one example dataset (control, GA = 27.14 weeks - same participant as Figure 1). Top row: the ADC and $\mathrm{T}{2}^{\ast }$ map from the joint fit. Bottom row: the $\mathrm{T}{2}^{\ast }$ map from the additional multi-echo gradient echo scan. The blue arrow shows the increased ADC and $\mathrm{T}{2}^{\ast }$ in the lobule centres, the pink arrow the increase on the chorionic plate and the orange arrows the increase in ADC and reduction in $\mathrm{T}{2}^{\ast }$ in the septa between the lobules.

Figure 3:

$\mathrm{T}{2}^{\ast }$ maps across gestation from joint $\mathrm{T}{2}^{\ast }$-diffusivity model fit to the protocol 1 combined $\mathrm{T}{2}^{\ast }$-diffusion data. Subplot titles give the gestational age in weeks and text color denotes the cohort.

Figure 4:

diffusivity maps across gestation from joint $\mathrm{T}{2}^{\ast }$-diffusivity model fit to the protocol 1 combined $\mathrm{T}{2}^{\ast }$-diffusion data.

Figure 5:

$\mathrm{T}{2}^{\ast }$ maps across gestation from joint $\mathrm{T}{2}^{\ast }$-diffusivity model fit to the protocol 2 combined $\mathrm{T}{2}^{\ast }$-diffusion data.

Figure 6:

diffusivity maps across gestation from joint $\mathrm{T}{2}^{\ast }$-diffusivity model fit to the protocol 2 combined $\mathrm{T}{2}^{\ast }$-diffusion data.

Figure 7:

Mean $\mathrm{T}{2}^{\ast }$ and ADC values over gestational age across the cohorts for protocol 1 (b=0, 50, 100, 150, 300, 500, 750, 1000 s/mm²) and protocol 2 (b=0, 20, 30, 50, 70, 100, 120 150, 300, 500, 750, 1000 s/mm²). Lines indicate the same participant scanned twice. Arrow in panel A indicates outlier scan with a clear contraction (see Discussion).

Figure 8:

Estimated $\mathrm{T}{2}^{\ast }$ (panel A) and ADC (panel B) values for both combined $\mathrm{T}{2}^{\ast }$-diffusion protocols.

Figure 9:

A) Mean $\mathrm{T}{2}^{\ast }$ values against gestation from monoexponential $\mathrm{T}{2}^{\ast }$ fit to standard multi-echo gradient echo scan. B) Mean placental $\mathrm{T}{2}^{\ast }$ values from joint model fit to combined $\mathrm{T}{2}^{\ast }$-diffusion scan (protocol 1) against mean placental $\mathrm{T}{2}^{\ast }$ values from fit to standard multi-echo gradient echo scan. C) As middle panel but for protocol 2.