AI Clinician XP2 - A Pilot Study of the AI Clinician Running in Real Time in the ICU
Overview
- Phase
- N/A
- Intervention
- Not specified
- Conditions
- Sepsis
- Sponsor
- Imperial College London
- Enrollment
- 64
- Locations
- 2
- Primary Endpoint
- Patient identification
- Status
- Not Yet Recruiting
- Last Updated
- 3 years ago
Overview
Brief Summary
The cornerstone of sepsis resuscitation is the administration of intravenous fluids (IVF) and/or vasopressors (drugs that squeeze the blood vessels to increase blood pressure) to maintain blood flow to prevent organ failure. However, there is huge uncertainty around the individual dosing of these drugs in an individual patient, partially due to high sepsis heterogeneity. The current guidelines provide recommendations at a population-level but fail to individualise the decisions. Wrong decisions lead to poorer outcomes and increased ICU-resource use. A tool to personalise these medications could improve patient survival.
The investigators have developed a new method to automatically and continuously review and recommend the correct dose of these medications to doctors, which was created using artificial intelligence (AI) techniques applied to large medical databases. The method used is called reinforcement learning, and we call the technology the "AI Clinician".
In the AI Clinician XP1, the investigators tested the safety of the AI Clinician when running in "shadow mode", i.e. in pseudonymised batches of patient data presented to off-duty ICU clinicians. This enabled the investigators to 1) develop methods and software to connect to real-time electronic health records (EHR); 2) check the safety of the algorithm when used in a contemporary UK ICU patient cohort.
In XP2, the AI Clinician will be running in real-time on dedicated computers at the bedside of actual patients in 4 ICUs across 2 NHS Trusts (Three ICUs at ICHT and one ICU at UCLH).
Detailed Description
Sepsis is life-threatening organ dysfunction due to severe infection and affects 250,000 patients annually in the UK (pre-COVID-19), of whom 48,000 die. In addition, virtually all COVID-19 intensive care unit (ICU) deaths had sepsis. It is a leading cause of death and the most expensive condition treated in hospitals. It was recognised as a top research priority by the James Lind Alliance, a partnership of patients and clinicians to prioritise the most pressing unanswered questions facing the NHS. The cornerstone of sepsis resuscitation is the administration of intravenous fluids (IVF) and/or vasopressors (drugs that squeeze the blood vessels to increase blood pressure) to maintain blood flow to prevent organ failure. However, there is huge uncertainty around the individual dosing of these drugs in an individual patient, partially due to high sepsis heterogeneity. The current guidelines provide recommendations at a population-level but fail to individualise the decisions. Wrong decisions lead to poorer outcomes and increased ICU-resource use. A tool to personalise these medications could improve patient survival. The investigators have developed a new method to automatically and continuously review and recommend the correct dose of these medications to doctors, which was created using artificial intelligence (AI) techniques applied to large medical databases. The method used is called reinforcement learning, and we call the technology the "AI Clinician". In the AI Clinician XP1, the investigators tested the safety of the AI Clinician when running in "shadow mode", i.e. in pseudonymised batches of patient data presented to off-duty ICU clinicians. This enabled the investigators to 1) develop methods and software to connect to real-time electronic health records (EHR); 2) check the safety of the algorithm when used in a contemporary UK ICU patient cohort. In XP2, the AI Clinician will be running in real-time on dedicated computers at the bedside of actual patients in 4 ICUs across 2 NHS Trusts (Three ICUs at ICHT and one ICU at UCLH). This present experiment will test the feasibility of running the AI Clinician in real-time in operational ICUs, in preparation for a future large scale multicentre randomised trial that will test for an improvement in clinically relevant outcomes. At this stage and in the interest of focusing on prescribers first, we will only be testing the use of the system by ICU doctors. Studies with nurses will be conducted in the future.
Investigators
Eligibility Criteria
Inclusion Criteria
- •For patients:
- •Adult \> 18yr
- •Admitted to an ICU in a participating centre
- •With early (within 24 of onset) sepsis (as defined by the sepsis-3 definition)
- •For full escalation (no ceiling of care, e.g. patient "not for vasopressors")
- •Expected to survive more than 24h
- •Has not opted-out for use of their data for research (NHS and NHS-X website)
- •For clinician participants:
- •ICU doctors at the senior registrar, ICU fellow or consultant level
Exclusion Criteria
- •For patients:
- •Not for full active care, e.g. not for vasopressors
- •Not expected to survive more than 24hr
- •Elective surgical admission (these patients are regularly on antibiotics but given as a prophylaxis, with no sepsis)
- •Opted-out for use of their data for research (NHS and NHS-X website)
- •For clinician participants:
- •Declined participation
Outcomes
Primary Outcomes
Patient identification
Time Frame: 6 months
Number of subjects identified and presented to a bedside doctor each week in each centre Number of times the system is used for each patient
System data
Time Frame: 6 months
System availability: success/failure of generating a response. Delay in generating response when the system is triggered. Number and nature of technical issues (drop-outs, freezes). An independent online form (survey-type) will be created to log all technical issues that the users may encounter (e.g., system unavailable, login issues etc). This survey will be kept on the same research laptop, but separate from the AI Clinician application, so it can't be affected by server outage for example. Server status, down-time events, planned and unplanned outages. These events can be monitored remotely and logged by the ICT team.
Usability data
Time Frame: 6 months
Data availability: what percentage of essential and optional data fields are available 24/7.
Anonymised patients' data
Time Frame: 6 months
Doses received of intravenous fluids and vasopressors , Urine output and fluid balance, Presence of sedation, mechanical ventilation, dialysis (binary)
Evaluators' data
Time Frame: 6 Months
Clinicians Gender, grade and seniority At each evaluation of the AI the database will capture and record the following: Agreement with AI suggested dose: does it appear appropriate, too high or too low? Will you modify your prescription based on the AI suggestion? (yes/no) Would you intervene if the AI dose was to be administered automatically? (yes/no).
Clinician Interviews
Time Frame: 6 Months
At the end of the study 2 participants will be qualitatively interviewed (with audio recording, for transcription +/- thematic analysis)