TIST-Net: style transfer in dynamic contrast enhanced MRI using spatial and temporal information.

Objective.Training deep learning models for image registration or segmentation of dynamic contrast enhanced (DCE) MRI data is challenging. This is mainly due to the wide variations in contrast enhancement within and between patients. To train a model effectively, a large dataset is needed, but acquiring it is expensive and time consuming. Instead, style transfer can be used to generate new images from existing images. In this study, our objective is to develop a style transfer method that incorporates spatio-temporal information to either add or remove contrast enhancement from an existing image.Approach.We propose a temporal image-to-image style transfer network (TIST-Net), consisting of an auto-encoder combined with convolutional long short-term memory networks. This enables disentanglement of the content and style latent spaces of the time series data, using spatio-temporal information to learn and predict key structures. To generate new images, we use deformable and adaptive convolutions which allow fine grained control over the combination of the content and style latent spaces. We evaluate our method, using popular metrics and a previously proposed contrast weighted structural similarity index measure. We also perform a clinical evaluation, where experts are asked to rank images generated by multiple methods.Main Results.Our model achieves state-of-the-art performance on three datasets (kidney, prostate and uterus) achieving an SSIM of 0.91 ± 0.03, 0.73 ± 0.04, 0.88 ± 0.04 respectively when performing style transfer between a non-enhanced image and a contrast-enhanced image. Similarly, SSIM results for style transfer from a contrast-enhanced image to a non-enhanced image were 0.89 ± 0.03, 0.82 ± 0.03, 0.87 ± 0.03. In the clinical evaluation, our method was ranked consistently higher than other approaches.Significance.TIST-Net can be used to generate new DCE-MRI data from existing images. In future, this may improve models for tasks such as image registration or segmentation by allowing small training datasets to be expanded.

UCP1 expression in human brown adipose tissue is inversely associated with cardiometabolic risk factors.

OBJECTIVE: Brown adipose tissue (BAT) is a therapeutic target for obesity. 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) is commonly used to quantify human BAT mass and activity. Detectable 18F-FDG uptake by BAT is associated with reduced prevalence of cardiometabolic disease. However, 18F-FDG uptake may not always be a reliable marker of BAT thermogenesis, for example, insulin resistance may reduce glucose uptake. Uncoupling protein 1 (UCP1) is the key thermogenic protein in BAT. Therefore, we hypothesised that UCP1 expression may be altered in individuals with cardiometabolic risk factors. METHODS: We quantified UCP1 expression as an alternative marker of thermogenic capacity in BAT and white adipose tissue (WAT) samples (n = 53) and in differentiated brown and white pre-adipocytes (n = 85). RESULTS: UCP1 expression in BAT, but not in WAT or brown/white differentiated pre-adipocytes, was reduced with increasing age, obesity, and adverse cardiometabolic risk factors such as fasting glucose, insulin, and blood pressure. However, UCP1 expression in BAT was preserved in obese subjects of <40 years of age. To determine if BAT activity was also preserved in vivo, we undertook a case-control study, performing 18F-FDG scanning during mild cold exposure in young (mean age ∼22 years) normal weight and obese volunteers. 18F-FDG uptake by BAT and BAT volume were similar between groups, despite increased insulin resistance. CONCLUSION: 18F-FDG uptake by BAT and UCP1 expression are preserved in young obese adults. Older subjects retain precursor cells with the capacity to form new thermogenic adipocytes. These data highlight the therapeutic potential of BAT mass expansion and activation in obesity.