Search

A nonoverlapping domain decomposition approach with uniform and matching grids is used to define and compute the orthogonal spline collocation solution of the Dirichlet boundary value problem for Poisson's equation on a square partitioned into four squares. The collocation solution on four interfaces is computed using the preconditioned conjugate gradient method with the preconditioner defined in terms of interface preconditioners for the adjacent squares. The collocation solution on four squares is computed by a matrix decomposition method that uses fast Fourier transforms. With the number of preconditioned conjugate gradient iterations proportional to log2 N, the total cost of the algorithm is O( N2 log2 N), where the number of unknowns in the collocation solution is O( N2). The approach presented in this paper, along with that in [B.Bialecki and M.Dryja, A nonoverlapping domain decomposition method for orthogonal spline collocation problems. SIAM J. Numer. Anal. (2003) 41, 1709 – 1728.], generalizes to variable coefficient equations on rectangular polygons partitioned into many subrectangles and is well suited for parallel computation. [ABSTRACT FROM AUTHOR]

We propose and analyse a fully discrete Petrov--Galerkin method with quadrature, for solving second-order, variable coefficient, elliptic boundary value problems on rectangular domains. In our scheme, the trial space consists of C² splines of degree r ≥ 3, the test space consists of C0 splines of degree r - 2, and we use composite (r - 1)-point Gauss quadrature. We show existence and uniqueness of the approximate solution and establish optimal order error bounds in H², H¹ and L² norms. [ABSTRACT FROM AUTHOR]

Objectives: Designing a framework representing radiology results in a standards-based data structure using joint Radiological Society of North America/American College of Radiology Common Data Elements (CDEs) as the semantic labels on standard structures. This allows radiologist-created report data to integrate with artificial intelligence-generated results for use throughout downstream systems., Materials and Methods: We developed a framework modeling radiology findings as Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) observations using CDE set/element identifiers as standardized semantic labels. This framework deploys CDE identifiers to specify radiology findings and attributes, providing consistent labels for radiology report concepts-diagnoses, recommendations, tabular/quantitative data-with built-in integration with RadLex, SNOMED CT, LOINC, and other ontologies. Observation structures fit within larger HL7 FHIR DiagnosticReport resources, providing output including both nuanced text and structured data., Results: Labeling radiology findings as discrete data for interchange between systems requires two components: structure and semantics. CDE definitions provide semantic identifiers for findings and their component values. The FHIR observation resource specifies a structure for associating identifiers with radiology findings in the context of reports, with CDE-encoded observations referring to definitions for CDE identifiers in a central repository. The discussion includes an example of encoding pulmonary nodules on a chest CT as CDE-labeled observations, demonstrating the application of this framework to exchange findings throughout the imaging workflow, making imaging data available to downstream clinical systems., Discussion: CDE-labeled observations establish a lingua franca for encoding, exchanging, and consuming radiology data at the level of individual findings, facilitating use throughout healthcare systems., Importance: CDE-labeled FHIR observation objects can increase the value of radiology results by facilitating their use throughout patient care., (© The Author(s) 2024. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing., (© 2022. The Author(s).)

Transmission of Imaging and Data (TRIAD) is a standard-based system built by the American College of Radiology to provide the seamless exchange of images and data for accreditation of clinical trials and registries. Scripts of structures' names validation profiles created in TRIAD are used in the automated submission process. It is essential for users to understand the logistics of these scripts for successful submission of radiation therapy cases with less iteration., (Copyright © 2016 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.)

The intent of this project was to use object storage and its database, which has the ability to add custom extensible metadata to an imaging object being stored within the system, to harness the power of its search capabilities, and to close the technology gap that healthcare faces. This creates a non-disruptive tool that can be used natively by both legacy systems and the healthcare systems of today which leverage more advanced storage technologies. The base infrastructure can be populated alongside current workflows without any interruption to the delivery of services. In certain use cases, this technology can be seen as a true alternative to the VNA (Vendor Neutral Archive) systems implemented by healthcare today. The scalability, security, and ability to process complex objects makes this more than just storage for image data and a commodity to be consumed by PACS (Picture Archiving and Communication System) and workstations. Object storage is a smart technology that can be leveraged to create vendor independence, standards compliance, and a data repository that can be mined for truly relevant content by adding additional context to search capabilities. This functionality can lead to efficiencies in workflow and a wealth of minable data to improve outcomes into the future.