COVID-19
Conditions
Brief summary
This is a pilot study to assess whether artificial intelligence (AI) combined with continuous vital signs monitoring from wearable sensors can predict clinically relevant outcomes in patients with suspected or confirmed Covid-19 infection on general medical wards.
Detailed description
Adult patients on general medical wards with COVID-19 infection considered to be at high risk of deterioration will be asked to wear vital signs sensors for the duration of their hospital stay. These sensors are an established method of recording patient vital signs and are CE marked. Patients enrolled in the study will continue to receive routine medical care as directed by their treating team. All data recorded from the wearable sensors in this study will be analysed in conjunction with routine data collected during the patient's treatment. Several models will be created using deep learning AI techniques with the aim of reliably predicting several important clinical outcomes. The study will identify whether continuous monitoring alone can improve identification of deteriorating patients compared to traditional vital signs and if the addition of AI technology / algorithms can provide even earlier identification.
Interventions
DEVICEContinuous vital sign monitoring - Isansys Patient Status Engine
CE marked wearable continuous vital signs monitors
OTHERMachine Learning/AI Algorithm
Patient data will be subjected to machine learning/AI algorithms to determine whether algorithms may be beneficial as an early indication of patient's condition worsening.
Sponsors
Manchester University NHS Foundation Trust
Aptus Clinical Ltd.
Zenzium Ltd.
The Christie NHS Foundation Trust
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
OTHER
Masking
NONE
Masking description
The treating team on the ward will be blinded to the observations recorded by the wearable vital signs sensors
Eligibility
Sex/Gender
ALL
Age
16 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Participants are eligible to be included in the study only if all of the following criteria apply: 1. Adult (aged 16 years or older), hospital inpatients 2. Suspected or confirmed COVID-19 infection (nasopharyngeal swab sent or planned): 1. Positive nasopharyngeal swab during this admission OR 2. Nasopharyngeal swab pending during this admission and the treating team suspect COVID-19 OR 3. Negative nasopharyngeal swab during this admission but the treating team continue to suspect COVID-19 OR 4. Positive nasopharyngeal swab in the last 7 days 3. Emergency admission to hospital within the last 72 hours and/or a positive nasopharyngeal test within the last 72 hours taken from a patient who was already an inpatient at the time the swab was taken. 4. Symptoms consistent with COVID-19 infection at the time of admission or when swab taken: cough, shortness of breath, alteration to sense of taste or smell, fevers or other symptoms in keeping with COVID-19 in the opinion of the study team. 5. For full active treatment (including escalation to critical care) 6. The patient is at risk of deterioration (as evidenced by a requirement for supplementary oxygen)
Exclusion criteria
Participants are excluded from the study if any of the following criteria apply: 1. Patients unable to give informed consent. 2. Patients with a life expectancy of \<24hours. 3. Known allergy or history of contact dermatitis to medical adhesives. 4. Patients with pacemakers, implantable defibrillators or neurostimulators. 5. Patients with an arterio-venous fistula in either arm.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Development of an AI model to predict clinically relevant outcomes for ward-based patients with COVID-19 monitored for up to 20 days. Metrics to be employed depend on the algorithm used but include, Log-Loss, precision and/or recall and confusion matrix. | 1 year |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Performance of the wearable vital signs sensor as measured by the percentage of possible data capture that is actually obtained | 1 year | — |
| Look for evidence of circadian disruption in the vital signs of the enrolled patients. | 1 year | To investigate whether circadian rhythm disruption is involved in COVID-19 |
Countries
United Kingdom
Outcome results
None listed