Lung2Lung: Volumetric Style Transfer with Self-Ensembling for High-Resolution Cross-Volume Computed Tomography

Chest computed tomography (CT) at inspiration is often complemented by an expiratory CT for identifying peripheral airways disease in the form of air trapping. Additionally, co-registered inspiratory-expiratory volumes are used to derive several clinically relevant measures of local lung function. Acquiring CT at different volumes, however, increases radiation dosage, acquisition time, and may not be achievable due to various complications, limiting the utility of registration-based measures, To address this, we propose Lung2Lung - a style-based generative adversarial approach for translating CT images from end-inspiratory to end-expiratory volume. Lung2Lung addresses several limitations of the traditional generative models including slicewise discontinuities, limited size of generated volumes, and their inability to model neural style at a volumetric level. We introduce multiview perceptual similarity (MEAL) to capture neural styles in 3D. To incorporate global information into the training process and refine the output of our model, we also propose self-ensembling (SE). Lung2Lung, with MEAL and SE, is able to generate large 3D volumes of size 320 x 320 x 320 that are validated using a diverse cohort of 1500 subjects with varying disease severity. The model shows superior performance against several state-of-the-art 2D and 3D generative models with a peak-signal-to-noise ratio of 24.53 dB and structural similarity of 0.904. Clinical validation shows that the synthetic volumes can be used to reliably extract several clinical endpoints of chronic obstructive pulmonary disease.

Single volume lung biomechanics from chest computed tomography using a mode preserving generative adversarial network

Local tissue expansion of the lungs is typically derived by registering computed tomography (CT) scans acquired at multiple lung volumes. However, acquiring multiple scans incurs increased radiation dose, time, and cost, and may not be possible in many cases, thus restricting the applicability of registration-based biomechanics. We propose a generative adversarial learning approach for estimating local tissue expansion directly from a single CT scan. The proposed framework was trained and evaluated on 2500 subjects from the SPIROMICS cohort. Once trained, the framework can be used as a registration-free method for predicting local tissue expansion. We evaluated model performance across varying degrees of disease severity and compared its performance with two image-to-image translation frameworks - UNet and Pix2Pix. Our model achieved an overall PSNR of 18.95 decibels, SSIM of 0.840, and Spearman's correlation of 0.61 at a high spatial resolution of 1 mm3.

Geodesic Density Regression for Correcting 4DCT Pulmonary Respiratory Motion Artifacts

Pulmonary respiratory motion artifacts are common in four-dimensional computed tomography (4DCT) of lungs and are caused by missing, duplicated, and misaligned image data. This paper presents a geodesic density regression (GDR) algorithm to correct motion artifacts in 4DCT by correcting artifacts in one breathing phase with artifact-free data from corresponding regions of other breathing phases. The GDR algorithm estimates an artifact-free lung template image and a smooth, dense, 4D (space plus time) vector field that deforms the template image to each breathing phase to produce an artifact-free 4DCT scan. Correspondences are estimated by accounting for the local tissue density change associated with air entering and leaving the lungs, and using binary artifact masks to exclude regions with artifacts from image regression. The artifact-free lung template image is generated by mapping the artifact-free regions of each phase volume to a common reference coordinate system using the estimated correspondences and then averaging. This procedure generates a fixed view of the lung with an improved signal-to-noise ratio. The GDR algorithm was evaluated and compared to a state-of-the-art geodesic intensity regression (GIR) algorithm using simulated CT time-series and 4DCT scans with clinically observed motion artifacts. The simulation shows that the GDR algorithm has achieved significantly more accurate Jacobian images and sharper template images, and is less sensitive to data dropout than the GIR algorithm. We also demonstrate that the GDR algorithm is more effective than the GIR algorithm for removing clinically observed motion artifacts in treatment planning 4DCT scans. Our code is freely available at https://github.com/Wei-Shao-Reg/GDR.