ABSTRACT
Machine Learning (ML) techniques in clinical decision support systems are scarce due to the limited availability of clinically validated and labelled training data sets. We present a framework to (1) enable quality controls at data submission toward ML appropriate data, (2) provide in-situ algorithm assessments, and (3) prepare dataframes for ML training and robust stochastic analysis. We developed and evaluated PiMS (Pandemic Intervention and Monitoring Systems): a remote monitoring solution for patients that are Covid-positive. The system was trialled at two hospitals in Melbourne, Australia (Alfred Health and Monash Health) involving 109 patients and 15 clinicians.
ABSTRACT
Purpose: As the COVID-19 emergency evolved, a wide range of 'new' technology based solutions were offered to meet clinical and occupational health needs in Europe. This technology extended beyond the standard medical devices usually deployed in clinical settings, and therefore required rapid assessment of suitability for use in hospitals. Here we describe a hospital-based COVID-19 technology assessment service (www.misa.ie/researchdevelopment/ bioengineering-lab/technology-assessment) that was developed and share our experience of its implementation. Material(s) and Method(s): A scientifically grounded assessment service was established to evaluate specific technological solutions. This service was led by a team of 2 Senior Medical Physicists and 1 Senior Clinical Engineer, with each assessment drawing on pan-hospital expertise and a specialist technology evaluation infrastructure. Each solution was evaluated using a standardized agile process: 1) user centric needs assessment;2) applicable literature and international standards review;3) balanced risk-benefit assessment;4) initial device functionality and usability assessment;5) in-depth device technical testing and safety assessment;6) rapid communications and detailed reporting;7) support for local clinical implementation/ installation with on-going evaluation. Evaluations were described in the form of short Bulletins with a webpage developed to share these findings internationally. Result(s): To date, a diverse range of technological systems and innovative solutions were evaluated, including thermal cameras for mass temperature screening, baby monitor devices for isolation room communications, augmented reality systems, a varied range of thermometers, and connected health technologies for remote working and clinical testing. Substantial variability in quality and standard of systems on offer was identified, with potential patient risks highlighted and mitigated. Critical success factors of the assessment service identified include: a central focus on the impact of solutions on both patients and staff, accessible local scientific and technical expertise supporting real-world testing and user feedback, an agile process which was responsive to high levels of uncertainty and a rapid communications process that was adaptive, responsive and connected both locally and nationally. Conclusion(s): Emergency situations, while challenging, are a huge stimulus for healthcare system-wide changes where barriers to technological innovation are significantly reduced, providing significant opportunities for adoption of new and innovative solutions. While there is a need for timely and practical technology assessments during an acute emergency, these should still be grounded in well-established scientific and safety principles that prioritize the health and safety of patients, staff and the public. A hospital-based COVID-19 technology assessment service has provided a practical and successful solution to this challenge.Copyright © 2023 Southern Society for Clinical Investigation.
ABSTRACT
Machine Learning (ML) techniques in clinical decision support systems are scarce due to the limited availability of clinically validated and labelled training data sets. We present a framework to (1) enable quality controls at data submission toward ML appropriate data, (2) provide in-situ algorithm assessments, and (3) prepare dataframes for ML training and robust stochastic analysis. We developed and evaluated PiMS (Pandemic Intervention and Monitoring Systems): a remote monitoring solution for patients that are Covid-positive. The system was trialled at two hospitals in Melbourne, Australia (Alfred Health and Monash Health) involving 109 patients and 15 clinicians. © 2022 IEEE.
ABSTRACT
The COVID-19 pandemic and the measures put in place to control its spread resulted in a collapse in global economic activity. Both governments and central banks responded to the deep economic crisis with unprecedented policy interventions. In this paper, we use both a global and a domestic structural macroeconomic model to estimate the impact of these exceptional policy supports on the international economy and on Ireland. Our analysis seeks to quantify the positive spillovers to the Irish economy from the fiscal, monetary and macroprudential policy interventions introduced globally and describes the transmission channels through which these policy actions affect Ireland. We also estimate the impact on economic activity and the labour market of the unprecedented fiscal packages introduced in Ireland. Our results indicate that the combination of international and domestic policy interventions have helped to substantially reduce the fall in output in Ireland from COVID-19, boosting growth in 2020 by up to 8 percentage points.