Artificial Intelligence Tools For Effective Monitoring Of Population At Distance During COVID-19 Pandemic. Results From An Italian Pilot Feasibility Study (Ricovai-19 Study).
Mazzanti, M.; Salvi, A.; Giacomini, S.; Perazzini, E.; Nitti, C.; Contucci, S.; D'Angelo, M.; Ballanti, D.; Ursino, D.; Marcosignori, M.; Romagnoli, R.; Tavio, M.; Giacometti, A.; Tomassetti, S.; Bonazzi, R.; Maccioni, M.; Zuccatosta, L.
Show abstract
In order to reduce the burden on healthcare systems and in particular to support an appropriate way to the Emergency Department (ED) access, home tele-monitoring patients was strongly recommended during the COVID-19 pandemic. Furthermore, paper from numerous groups has shown the potential of using data from wearable devices to characterize each individuals unique baseline, identify deviations from that baseline suggestive of a viral infection, and to aggregate that data to better inform population surveillance trends. However, no evidence about usage of Artificial Intelligence (AI) applicatives on digitally data collected from patients and doctors exists. With a growing global population of connected wearable users, this could potentially help to improve the earlier diagnosis and management of infectious individuals and improving timeliness and precision of tracking infectious disease outbreaks. During the study RICOVAI-19 (RICOVero ospedaliero con strumenti di Artificial Intelligence nei pazienti con COVid-19) performed in the Marche Region, Italy, we evaluated 129 subjects monitored at home in a six-months period between March 22, 2021 and October 22, 2021. During the monitoring, personal on demand health technologies were used to collect clinical and vital data in order to feed the database and the machine learning engine. The AI output resulted in a clinical stability index (CSI) which enables the system to deliver suggestions to the population and doctors about how intervene. Results showed the beneficial influence of CSI for predicting clinical classes of subjects and identifying who of them need to be admitted at ED. The same pattern of results was confirming the alert included in the decision support system in order to request further testing or clinical information in some cases. In conclusion, our study does support a high impact of AI tools on COVID-19 outcomes to fight this pandemic by driving new approaches to public awareness.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Using wearable and nearable devices in telerehabilitation for COPD: A review of digital endpoints in home-based programs 94%
- An automated Dashboard to improve laboratory COVID-19 diagnostics management 94%
- Development and Validation of a Machine Learning Model Integrated with the Clinical Workflow for Inpatient Discharge Date Prediction 93%
Similar papers in this journal
- A Cross-sectional Feasibility Study to Evaluate the Usability and Efficacy of “Swaasa”: An AI-Platform for Rapid Respiratory Health Assessment 95%
- Development and Clinical Validation of Swaasa AI Platform for screening and prioritization of Pulmonary TB 94%
- A multipurpose machine learning approach to predict COVID-19 negative prognosis in Sao Paulo, Brazil 93%
Similar papers in this journal
- Automated Image Transcription for Perinatal Blood Pressure Monitoring Using Mobile Health Technology 95%
- From theoretical models to practical deployment: A perspective and case study of opportunities and challenges in AI-driven healthcare research for low-income settings 94%
- Developing and validating an explainable digital mortality prediction tool for extremely preterm infants 94%
Similar papers in this journal
- Towards a COVID-19 symptom triad: The importance of symptom constellations in the SARS-CoV-2 pandemic 95%
- Validity and reliability of an app-based medical device to empower individuals in evaluating their physical capacities 94%
- REPLICCAR II Study: Data Quality Audit in the Paulista Cardiovascular Surgery Registry 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.