Automatic quantitative computed tomography measurement of longitudinal lung volume loss in interstitial lung diseases.

OBJECTIVES: To compare the lung CT volume (CTvol) and pulmonary function tests in an interstitial lung disease (ILD) population. Then to evaluate the CTvol loss between idiopathic pulmonary fibrosis (IPF) and non-IPF and explore a prognostic value of annual CTvol loss in IPF. METHODS: We conducted in an expert center a retrospective study between 2005 and 2018 on consecutive patients with ILD. CTvol was measured automatically using commercial software based on a deep learning algorithm. In the first group, Spearman correlation coefficients (r) between forced vital capacity (FVC), total lung capacity (TLC), and CTvol were calculated. In a second group, annual CTvol loss was calculated using linear regression analysis and compared with the Mann-Whitney test. In a last group of IPF patients, annual CTvol loss was calculated between baseline and 1-year CTs for investigating with the Youden index a prognostic value of major adverse event at 3 years. Univariate and log-rank tests were calculated. RESULTS: In total, 560 patients (4610 CTs) were analyzed. For 1171 CTs, CTvol was correlated with FVC (r: 0.86) and TLC (r: 0.84) (p < 0.0001). In 408 patients (3332 CT), median annual CTvol loss was 155.7 mL in IPF versus 50.7 mL in non-IPF (p < 0.0001) over 5.03 years. In 73 IPF patients, a relative annual CTvol loss of 7.9% was associated with major adverse events (log-rank, p < 0.0001) in univariate analysis (p < 0.001). CONCLUSIONS: Automated lung CT volume may be an alternative or a complementary biomarker to pulmonary function tests for the assessment of lung volume loss in ILD. KEY POINTS: • There is a good correlation between lung CT volume and forced vital capacity, as well as for with total lung capacity measurements (r of 0.86 and 0.84 respectively, p < 0.0001). • Median annual CT volume loss is significantly higher in patients with idiopathic pulmonary fibrosis than in patients with other fibrotic interstitial lung diseases (155.7 versus 50.7 mL, p < 0.0001). • In idiopathic pulmonary fibrosis, a relative annual CT volume loss higher than 9.4% is associated with a significantly reduced mean survival time at 2.0 years versus 2.8 years (log-rank, p < 0.0001).

A conventional-to-spectral CT image translation augmentation workflow for robust contrast injection-independent organ segmentation.

PURPOSE: In computed tomography (CT) cardiovascular imaging, the numerous contrast injection protocols used to enhance structures make it difficult to gather training datasets for deep learning applications supporting diverse protocols. Moreover, creating annotations on noncontrast scans is extremely tedious. Recently, spectral CT's virtual-noncontrast images (VNC) have been used as data augmentation to train segmentation networks performing on enhanced and true-noncontrast (TNC) scans alike, while improving results on protocols absent of their training dataset. However, spectral data are not widely available, making it difficult to gather specific datasets for each task. As a solution, we present a data augmentation workflow based on a trained image translation network, to bring spectral-like augmentation to any conventional CT dataset. METHOD: The conventional CT-to-spectral image translation network (HUSpectNet) was first trained to generate VNC from conventional housnfied units images (HU), using an unannotated spectral dataset of 1830 patients. It was then tested on a second dataset of 300 spectral CT scans by comparing VNC generated through deep learning (VNCDL ) to their true counterparts. To illustrate and compare our workflow's efficiency with true spectral augmentation, HUSpectNet was applied to a third dataset of 112 spectral scans to generate VNCDL along HU and VNC images. Three different three-dimensional (3D) networks (U-Net, X-Net, and U-Net++) were trained for multilabel heart segmentation, following four augmentation strategies. As baselines, trainings were performed on contrasted images without (HUonly) and with conventional gray-values augmentation (HUaug). Then, the same networks were trained using a proportion of contrasted and VNC/VNCDL images (TrueSpec/GenSpec). Each training strategy applied to each architecture was evaluated using Dice coefficients on a fourth multicentric multivendor single-energy CT dataset of 121 patients, including different contrast injection protocols and unenhanced scans. The U-Net++ results were further explored with distance metrics on every label. RESULTS: Tested on 300 full scans, our HUSpectNet translation network shows a mean absolute error of 6.70 ± 2.83 HU between VNCDL and VNC, while peak signal-to-noise ratio reaches 43.89 dB. GenSpec and TrueSpec show very close results regardless of the protocol and used architecture: mean Dice coefficients (DSCmean ) are equal with a margin of 0.006, ranging from 0.879 to 0.938. Their performances significantly increase on TNC scans (p-values < 0.017 for all architectures) compared to HUonly and HUaug, with DSCmean of 0.448/0.770/0.879/0.885 for HUonly/HUaug/TrueSpec/GenSpec using the U-Net++ architecture. Significant improvements are also noted for all architectures on chest-abdominal-pelvic scans (p-values < 0.007) compared to HUonly and for pulmonary embolism scans (p-values < 0.039) compared to HUaug. Using U-Net++, DSCmean reaches 0.892/0.901/0.903 for HUonly/TrueSpec/GenSpec on pulmonary embolism scans and 0.872/0.896/0.896 for HUonly/TrueSpec/GenSpec on chest-abdominal-pelvic scans. CONCLUSION: Using the proposed workflow, we trained versatile heart segmentation networks on a dataset of conventional enhanced CT scans, providing robust predictions on both enhanced scans with different contrast injection protocols and TNC scans. The performances obtained were not significantly inferior to training the model on a genuine spectral CT dataset, regardless of the architecture implemented. Using a general-purpose conventional-to-spectral CT translation network as data augmentation could therefore contribute to reducing data collection and annotation requirements for machine learning-based CT studies, while extending their range of application.

Spectral augmentation for heart chambers segmentation on conventional contrasted and unenhanced CT scans: an in-depth study.

PURPOSE: Recently, machine learning has outperformed established tools for automated segmentation in medical imaging. However, segmentation of cardiac chambers still proves challenging due to the variety of contrast agent injection protocols used in clinical practice, inducing disparities of contrast between cavities. Hence, training a generalist network requires large training datasets representative of these protocols. Furthermore, segmentation on unenhanced CT scans is further hindered by the challenge of obtaining ground truths from these images. Newly available spectral CT scanners allow innovative image reconstructions such as virtual non-contrast (VNC) imaging, mimicking non-contrasted conventional CT studies from a contrasted scan. Recent publications have demonstrated that networks can be trained using VNC to segment contrasted and unenhanced conventional CT scans to reduce annotated data requirements and the need for annotations on unenhanced scans. We propose an extensive evaluation of this statement. METHOD: We undertake multiple trainings of a 3D multi-label heart segmentation network with (HU-VNC) and without (HUonly) VNC as augmentation, using decreasing training dataset sizes (114, 76, 57, 38, 29, 19 patients). At each step, both networks are tested on a multi-vendor, multi-centric dataset of 122 patients, including different protocols: pulmonary embolism (PE), chest-abdomen-pelvis (CAP), heart CT angiography (CTA) and true non-contrast scans (TNC). An in-depth comparison of resulting Dice coefficients and distance metrics is performed for the networks trained on the largest dataset. RESULTS: HU-VNC-trained on 57 patients significantly outperforms HUonly trained on 114 regarding CAP and TNC scans (mean Dice coefficients of 0.881/0.835 and 0.882/0.416, respectively). When trained on the largest dataset, significant improvements in all labels are noted for TNC and CAP scans (mean Dice coefficient of 0.882/0.416 and 0.891/0.835, respectively). CONCLUSION: Adding VNC images as training augmentation allows the network to perform on unenhanced scans and improves segmentations on other imaging protocols, while using a reduced training dataset.