Serum Potassium Prediction Using Machine Learning and Single-lead ECG

Mass General Brigham

Status: Withdrawn

Conditions: Chronic Kidney Diseases; Asthma; Infection; Anticoagulants; Increased; Chronic Obstructive Pulmonary Disease; Atrial Fibrillation Rapid; Hypertensive Urgency; Heart Failure; Gout Flare

Treatments: Other: Potassium estimation algorithm

Study type: Observational

Funder types: Other; Industry

Identifiers

NCT07493798

2017P002583c

Details and patient eligibility

About

This is a retrospective study drawing on data from the Brigham and Women's Hospital Home Hospital Program's Database. Sociodemographic and clinical data from a training cohort were used to train a machine learning algorithm to predict blood potassium throughout a patient's admission. This algorithm was then validated in a validation cohort.

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion and exclusion criteria

Was a subject in the Brigham and Women's Home Hospital study and has a completed record in the study's database.

Trial design

0 participants in 2 patient groups

Training

Description:

A subset of patients that are used to train the machine learning algorithm.

Treatment:

Other: Potassium estimation algorithm

Validation

Description:

A subset of patients that are "held back" and used to validate the algorithm's accuracy.

Treatment:

Other: Potassium estimation algorithm