arXiv subjects

Chris Marshall

Publications and source records attributed to Chris Marshall.

3 recordsLinked to original sources

arXiv2025

Horizon Scans can be accelerated using novel information retrieval and artificial intelligence tools

Introduction: Horizon scanning in healthcare assesses early signals of innovation, crucial for timely adoption. Current horizon scanning faces challenges in efficient information retrieval and analysis, especially from unstructured sources like news, presenting a need for innovative tools. Methodology: The study introduces SCANAR and AIDOC, open-source Python-based tools designed to improve horizon scanning. SCANAR automates the retrieval and processing of news articles, offering functionalities such as de-duplication and unsupervised relevancy ranking. AIDOC aids filtration by leveraging AI to reorder textual data based on relevancy, employing neural networks for semantic similarity, and subsequently prioritizing likely relevant entries for human review. Results: Twelve internal datasets from horizon scans and four external benchmarking datasets were used. SCANAR improved retrieval efficiency by automating processes previously dependent on manual labour. AIDOC displayed work-saving potential, achieving around 62% reduction in manual review efforts at 95% recall. Comparative analysis with benchmarking data showed AIDOC's performance was similar to existing systematic review automation tools, though performance varied depending on dataset characteristics. A smaller case-study on our news datasets shows the potential of ensembling large language models within the active-learning process for faster detection of relevant articles across news datasets. Conclusion: The validation indicates that SCANAR and AIDOC show potential to enhance horizon scanning efficiency by streamlining data retrieval and prioritisation. These tools may alleviate methodological limitations and allow broader, swifter horizon scans. Further studies are suggested to optimize these models and to design new workflows and validation processes that integrate large language models.

cs.IR↗

arXiv2023

NEMA NU 2-2018 performance evaluation of a new generation digital 32-cm axial field-of-view Omni Legend PET-CT

A NEMA performance evaluation was conducted on the new General Electric (GE) digital Omni Legend PET-CT system with 32-cm extended field-of-view. This study marks the introduction of the first-ever commercially available clinical digital bismuth germanate technology. Testing was performed in accordance with the NEMA NU2-2018 standard. A comparison was made with the performance of two other commercial GE scanners with extended fields-of-view. A digital lutetium yttrium orthosilicate system (Discovery MI - 6 ring) and a non-digital bismuth germanate system (Discovery IQ). For the Omni assessment, the tangential, radial, and axial spatial resolutions at 1 cm radial offset were measured as 3.76 mm, 3.73 mm, and 4.25 mm FWHM. The total system sensitivity to a line source at the center was 44.36 cps/kBq. The peak NECR was 501 kcps at 17.8 kBq/mL. The scatter fraction at NECR peak was 35.48%, and the maximum count-rate error at and below NEC peak was 5.5%. Sphere contrast recovery coefficients were from 52% (10 mm) to 93% (37 mm). The system does not use time of flight; thus, no assessment of timing resolution was made. The PET-CT co-registration accuracy was 2.4 mm. The performance of the Omni Legend surpassed that of the Discovery MI on all NEMA tests, except for assessments of background variability (image noise). Time of flight is associated with inherent improvements in signal-to-noise ratio. In lieu of time of flight capabilities, the Omni provides software corrections in the form of a pre-trained neural network (trained on non-ToF to ToF). With such corrections, average performance is competitive when compared to ToF systems. Further validation is required to optimize clinical imaging protocols and hyperparameters associated with such software corrections and to examine the effect of non-linear corrections with varying target size, particularly for real world, clinical scans.

physics.med-ph↗

arXiv2018

Progress in measurements of 0.1--10 GeV neutrino-nucleus scattering and anticipated results from future experiments

Neutrino interactions with nuclei have been the subject of intense interest over the last 15 years. Current and future measurements of neutrino oscillation and exotic physics use order 0.1--10 GeV neutrinos on a range of nuclear targets ($^{12}$C, $^{16}$O, $^{40}$Ar). As the precision of these experiments has increased, information from their detectors and dedicated experiments indicate deficiencies in the modeling of neutrino interactions on nuclear targets. Here, we present the current state of knowledge about neutrino-nucleus interactions, the challenge of extracting the cross section of these processes, and current experimental puzzles in the field. We also look forward to new and novel measurements and efforts in the future which seek to resolve these questions.

hep-ex↗