Abdominal Sepsis, Clinical Deterioration, Hemodynamic Instability, Infections, Sepsis
Conditions
Keywords
sepsis, surgical ward, artificial intelligence, decision-support, continuous monitoring, wearable devices
Brief summary
The goal of this prospective observational study is to develop and utilize an Artificial Intelligence (AI) model for the prediction of postoperative sepsis in patients undergoing abdominal surgery. The main questions it aims to answer are: 1. Can a remote AI-driven monitoring system accurately predict sepsis risk in postoperative patients? 2. How effectively can this system integrate and analyze multimodal data for early sepsis detection in the surgical ward? Participants are equipped with non-invasive PPG-based wearable devices to continuously monitor vital signs and collect high-quality clinical data. This data, along with demographic and laboratory information from the Electronic Health Record (EHR) of the hospital, are used for AI model development and validation.
Interventions
DEVICEPRISM Tool
The intervention in this study involves an AI-driven clinical decision-support system, PRISM Tool, designed for the early prediction of sepsis in patients undergoing abdominal surgery. PRISM Tool integrates data from PPG-based wearable wireless devices that monitor vital signs, electronic health records, and laboratory tests. The AI model analyzes this multimodal data to proactively identify signs of sepsis providing an early warning score to clinicians. The distinguishing feature of this intervention is its use of real-time data and advanced AI analytics to enhance early sepsis detection, aiming to improve patient outcomes in postoperative care.
Sponsors
Aisthesis Medical P.C.
Larissa University Hospital
Technical University of Crete
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to 120 Years
Inclusion criteria
* Patients undergoing elective abdominal surgery. * Postoperative admission to the surgical ward. * Age 18 years or older, who are able and willing to participate and have given written consent. * On admission, the primary investigator assess their risk to deteriorate during the first 72 hours after admission as reasonably high.
Exclusion criteria
* \<18 years of age Known allergy or contraindication to the monitoring devices. * Pre-existing conditions that could interfere with the study (e.g., chronic sepsis, immunodeficiency disorders). * Day case surgery. * Pregnancy. * Immediate transfer to ICU postoperatively. * Patient refusal or unable to give written consent.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Accuracy of AI-Driven Sepsis Prediction in Postoperative Period | The accuracy of sepsis prediction will be assessed from the day of surgery, assessed daily for up to 7 days post-surgery or until hospital discharge. | This primary outcome measure evaluates the accuracy of an AI-driven monitoring system in predicting postoperative sepsis among patients undergoing abdominal surgery. The measure focuses on the system's ability to correctly identify sepsis, considering sensitivity, specificity, and predictive values. |
Countries
Greece
Contacts
STUDY_CHAIREleni Arnaoutoglou, MD, PhD
Larissa University Hospital
Outcome results
None listed