Machine Learning Platforms to Predict 30-day Mortality After Emergency Laparotomy

NCT05828914 | Unknown

• 1 other identifier: UW 22-334
• Study Type: observational
• Target: 5,000
• Locations: 1 country
• Sites: 1

Brief Summary

This study seeks to utilise retrospective patient data to train machine learning algorithms to predict the short term mortality and morbidity after an emergency laparotomy. Data will be collected via the Electronic Health records system at the Queen Mary Hospital Hong Kong. Machine learning models will be compared and the best-performing one will be explored for further optimization and deployment. Upon completion, we hope that this platform will aid clinicians to identify high risk patients and aid clinical decisions and peri-operative planning, with the aim to reduce mortality and morbidity in this high risk procedure.

Conditions

Mortality Rate

Outcome Measures

Primary Outcomes (1)

• Short term (30-day) mortality rate in emergency laparotomies and accuracy of NELA risk calculator in the Hong Kong population.

  We hypothesize that the machine learning algorithm with incorporation of the frailty will perform better than the existing NELA risk calculator

  January 2017 and April 2021

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

All patients that meet the eligibility criteria within the timeframe of observation will be included in this study for machine learning modeling. Hence, no sample size calculation is required for this study.

You may qualify if:

• Adult patients (\> 18 years old) undergoing EL in Queen Mary Hospital between January 2017 and April 2021 will be included in this study

You may not qualify if:

• Patients undergoing other emergency general surgical procedures (eg. laparoscopies, cholecystectomy, appendicectomy) will be excluded from this study.

Sponsors & Collaborators

• The University of Hong Kong: lead

Study Sites (1)

HKU Li Ka Shing Faculty of Medicine

Hong Kong, Guangdong, 999077, China

Location

Study Officials

• Michael Garnet Irwin, M.B. Ch.B

  The University of Hong Kong

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Michael Garnet Irwin, M.B. Ch.B

CONTACT

97018342; mgirwin@hku.hk

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor

Study Record Dates

• First Submitted: April 12, 2023
• First Posted: April 25, 2023
• Study Start: April 28, 2023
• Primary Completion: July 31, 2024
• Study Completion: July 31, 2025
• Last Updated: April 25, 2023

Record last verified: 2023-04

Locations

CN (1)