Free breathing lung T1 mapping using image registration in patients with idiopathic pulmonary fibrosis

Purpose

To assess the use of image registration for correcting respiratory motion in free breathing lung T1 mapping acquisition in patients with idiopathic pulmonary fibrosis (IPF).

Theory and Methods

The method presented used image registration to synthetic images during postprocessing to remove respiratory motion. Synthetic images were generated from a model of the inversion recovery signal of the acquired images that incorporated a periodic lung motion model. Ten healthy volunteers and 19 patients with IPF underwent 2D Look‐Locker T1 mapping acquisition at 1.5T during inspiratory breath‐hold and free breathing. Eight healthy volunteers and seven patients with IPF underwent T1 mapping acquisition during expiratory breath‐hold. Fourteen patients had follow‐up scanning at 6 months. Dice similarity coefficient (DSC) was used to evaluate registration efficacy.

Results

Image registration increased image DSC (P < .001) in the free breathing inversion recovery images. Lung T1 measured during a free breathing acquisition was lower in patients with IPF when compared with healthy controls (inspiration: P = .238; expiration: P = .261; free breathing: P = .021). Measured lung T1 was higher in expiration breath‐hold than inspiration breath‐hold in healthy volunteers (P < .001) but not in patients with IPF (P = .645). There were no other significant differences between lung T1 values within subject groups.

Conclusions

The registration technique significantly reduced motion in the Look‐Locker images acquired during free breathing and may improve the robustness of lung T1 mapping in patients who struggle to hold their breath. Lung T1 measured during a free breathing acquisition was significantly lower in patients with IPF when compared with healthy controls.