Self-trained deep convolutional neural network for noise reduction in CT.

Purpose: Supervised deep convolutional neural network (CNN)-based methods have been actively used in clinical CT to reduce image noise. The networks of these methods are typically trained using paired high- and low-quality data from a massive number of patients and/or phantom images. This training process is tedious, and the network trained under a given condition may not be generalizable to patient images acquired and reconstructed under different conditions. We propose a self-trained deep CNN (ST_CNN) method for noise reduction in CT that does not rely on pre-existing training datasets. Approach: The ST_CNN training was accomplished using extensive data augmentation in the projection domain, and the inference was applied to the data itself. Specifically, multiple independent noise insertions were applied to the original patient projection data to generate multiple realizations of low-quality projection data. Then, rotation augmentation was adopted for both the original and low-quality projection data by applying the rotation angle directly on the projection data so that images were rotated at arbitrary angles without introducing additional bias. A large number of paired low- and high-quality images from the same patient were reconstructed and paired for training the ST_CNN model. Results: No significant difference was found between the ST_CNN and conventional CNN models in terms of the peak signal-to-noise ratio and structural similarity index measure. The ST_CNN model outperformed the conventional CNN model in terms of noise texture and homogeneity in liver parenchyma as well as better subjective visualization of liver lesions. The ST_CNN may sacrifice the sharpness of vessels slightly compared to the conventional CNN model but without affecting the visibility of peripheral vessels or diagnosis of vascular pathology. Conclusions: The proposed ST_CNN method trained from the data itself may achieve similar image quality in comparison with conventional deep CNN denoising methods pre-trained on external datasets.

Professionally- and Self-Trained Service Dogs: Benefits and Challenges for Partners With Disabilities.

It has been widely reported that service dogs offer benefits to their human partners, however, it is unclear whether the expanding methods of training and roles of service dogs for their partners with various disabilities also provide similar benefits. This study aimed to investigate the self-reported experience of service dog partners to understand whether three different factors influence the benefits and drawbacks associated with partnering with a service dog: (1) different methods of training service dogs; (2) different severities of human partners' disabilities; (3) different roles of service dogs. Partners of service dogs were recruited to the web survey through service dog facilities and networking groups. Answers from 19 men and 147 women participants (91.8% living in the U.S.) were analyzed in this study. Participants experienced the expected benefits of service dogs, including increased independence, social relationships, self-esteem, and life satisfaction, and decreased anxiety, stress, and loneliness. However, the perceived benefits, concerns, and burdens differed depending on the partners' disabilities and the training history of the dogs. When first living with their service dogs, people who had self-trained their service dogs experienced more burdens than those living with professionally trained service dogs. No major reduction in expenses for assistance after acquiring a dog was reported. Personalized team training based on each person's disabilities and situation is required to optimize the benefits and minimize the burdens and concerns of living with service dogs.