Multiorgan structures detection using deep convolutional neural networks

Effect of Common Medications on Longitudinal Pectoralis Muscle Area in Smokers

Background

Cigarette smoke contributes to skeletal muscle wasting. While exercise and nutritional therapies are effective in improving skeletal muscle quantity and quality, the effect of medications on longitudinal muscle loss is unclear. We investigated whether long-term use of common medications affects longitudinal skeletal muscle changes in current and former smokers.

Methods

Using quantitative computed tomography imaging, we measured the 5-year changes in pectoralis muscle area (delta-PMA) and pectoralis muscle density (delta-PMD) of 4191 participants in the COPD Genetic Epidemiology (COPDGene®) study. We tested whether specific medications were associated with delta-PMA and/or delta-PMD using regression analyses. Propensity score matching (PSM) analysis was performed to determine the effect of the medications on longitudinal changes on delta-PMA.

Results

Over the study period, the median delta-PMA for the entire population showed a decrease of 2.23cm2 (interquartile range: -6.52, 1.54). Regression analyses demonstrated statin use was associated with less loss of PMA, whereas, aspirin use was associated with a greater loss of PMA. Specifically, in the PSM-adjusted analysis, statin use was associated with attenuated loss of PMA (median; -1.5 versus -2.5cm2, p=0.017), while aspirin use was associated with increased loss of PMA (median; -2.5 versus -1.9cm2, p=0.040).

Conclusion

In current and former smokers, statin use was associated with reduced pectoralis muscle wasting, while aspirin use was associated with increased muscle loss. Additional research is needed to verify these findings. (Clinicaltrials.gov identifier NCT00608764).

4D-CT deformable image registration using multiscale unsupervised deep learning

Deformable image registration (DIR) of 4D-CT images is important in multiple radiation therapy applications including motion tracking of soft tissue or fiducial markers, target definition, image fusion, dose accumulation and treatment response evaluations. It is very challenging to accurately and quickly register 4D-CT abdominal images due to its large appearance variances and bulky sizes. In this study, we proposed an accurate and fast multi-scale DIR network (MS-DIRNet) for abdominal 4D-CT registration. MS-DIRNet consists of a global network (GlobalNet) and local network (LocalNet). GlobalNet was trained using down-sampled whole image volumes while LocalNet was trained using sampled image patches. MS-DIRNet consists of a generator and a discriminator. The generator was trained to directly predict a deformation vector field (DVF) based on the moving and target images. The generator was implemented using convolutional neural networks with multiple attention gates. The discriminator was trained to differentiate the deformed images from the target images to provide additional DVF regularization. The loss function of MS-DIRNet includes three parts which are image similarity loss, adversarial loss and DVF regularization loss. The MS-DIRNet was trained in a completely unsupervised manner meaning that ground truth DVFs are not needed. Different from traditional DIRs that calculate DVF iteratively, MS-DIRNet is able to calculate the final DVF in a single forward prediction which could significantly expedite the DIR process. The MS-DIRNet was trained and tested on 25 patients' 4D-CT datasets using five-fold cross validation. For registration accuracy evaluation, target registration errors (TREs) of MS-DIRNet were compared to clinically used software. Our results showed that the MS-DIRNet with an average TRE of 1.2 ± 0.8 mm outperformed the commercial software with an average TRE of 2.5 ± 0.8 mm in 4D-CT abdominal DIR, demonstrating the superior performance of our method in fiducial marker tracking and overall soft tissue alignment.