Abstract

Abstract
Cardiovascular imaging studies provide a multitude of structural and functional data to better understand disease mechanisms. While pooling data across studies enables more powerful and broader applications, performing quantitative comparisons across datasets with varying acquisition or analysis methods is problematic due to inherent measurement biases specific to each protocol. We show how dynamic time warping and partial least squares regression can be applied to effectively map between left ventricular geometries derived from different imaging modalities and analysis protocols to account for such differences. To demonstrate this method, paired real-time 3D echocardiography (3DE) and cardiac magnetic resonance (CMR) sequences from 138 subjects were used to construct a mapping function between the two modalities to correct for biases in left ventricular clinical cardiac indices, as well as regional shape. Leave-one-out cross-validation revealed a significant reduction in mean bias, narrower limits of agreement, and higher intraclass correlation coefficients for all functional indices between CMR and 3DE geometries after spatiotemporal mapping. Meanwhile, average root mean squared errors between surface coordinates of 3DE and CMR geometries across the cardiac cycle decreased from 7 +/- 1 to 4 +/- 1 mm for the total study population. Our generalised method for mapping between time-varying cardiac geometries obtained using different acquisition and analysis protocols enables the pooling of data between modalities and the potential for smaller studies to leverage large population databases for quantitative comparisons.

Abstract
Continuous Integration and Continuous Deployment (CI/CD) pipelines play a vital role in the DevOps process, enabling developers to automate and enhance software delivery. However, the existence of multiple technologies, such as GitHub Actions, GitLab CI/CD, or Jenkins, poses challenges due to their lack of interoperability and the use of different programming languages for pipeline construction. To address these challenges and improve the CI/CD process, our objective is to develop a block-based language specifically designed for representing CI/CD pipelines. With our language, we intend to empower users to more easily create correct pipelines. Through an interactive and user-friendly process, our approach guides users in constructing pipelines, ensuring accuracy and reducing errors. Additionally, our language will facilitate seamless transitions between different pipeline technologies, providing users with flexibility and ease of adoption.

Abstract
Medical imaging, mainly Magnetic Resonance Imaging (MRI), plays a predominant role in healthcare diagnosis. Nevertheless, the diagnostic process is prone to errors and is conditioned by available medical data, which might be insufficient. A novel solution is resorting to image generation algorithms to address these challenges. Thus, this paper presents a Deep Learning model based on a Deep Convolutional Generative Adversarial Network (DCGAN) architecture. Our model generates 2D MRI images of size 256x256, containing an axial view of the brain with a tumor. The model was implemented using ChainerMN, a scalable and flexible framework that enables faster and parallel training of Deep Learning networks. The images obtained provide an overall representation of the brain structure and the tumoral area and show considerable brain-tumor separation. For this purpose, and owing to their previous state-of-the-art results in general image-generation tasks, we conclude that GAN-based models are a promising approach for medical imaging.

Abstract

Abstract
The self-repair capacity of human tissue is limited, motivating the arising of tissue engineering (TE) in building temporary scaffolds that envisage the regeneration of human tissues, including articular cartilage. However, despite the large number of preclinical data available, current therapies are not yet capable of fully restoring the entire healthy structure and function on this tissue when significantly damaged. For this reason, new biomaterial approaches are needed, and the present work proposes the development and characterization of innovative polymeric membranes formed by blending marine origin polymers, in a chemical free cross-linking approach, as biomaterials for tissue regeneration. The results confirmed the production of polyelectrolyte complexes molded as membranes, with structural stability resulting from natural intermolecular interactions between the marine biopolymers collagen, chitosan and fucoidan. Furthermore, the polymeric membranes presented adequate swelling ability without compromising cohesiveness (between 300 and 600%), appropriate surface properties, revealing mechanical properties similar to native articular cartilage. From the different formulations studied, the ones performing better were the ones produced with 3 % shark collagen, 3% chitosan and 10% fucoidan, as well as with 5% jellyfish collagen, 3% shark collagen, 3% chitosan and 10% fucoidan. Overall, the novel marine polymeric membranes demonstrated to have promising chemical, and physical properties for tissue engineering approaches, namely as thin biomaterial that can be applied over the damaged articular cartilage aiming its regeneration.