Early Detection of Clinical Deterioration in Patients With COVID-19 Using Machine Learning (COVID-19)

The aim of this study is to use artificial intelligence in the form of machine learning analysing vital signs as well as symptoms of patients suffering from Covid19 to identify predictors of disease progression and severe course of disease.

Study Overview

• Status: Unknown
• Conditions: Covid19
• Intervention / Treatment: Other: Machine learning; Other: Machine based evaluation

• Study Type: Observational
• Enrollment (Anticipated): 1000

Contacts and Locations

• Study Contact: Name: Annika Buchholz, PhD; Phone Number: +49 151 51819576; Email: annika.buchholz@tuebingen.mpg.de

Study Locations

• Germany
  • Tuebingen, Germany, 72076
    • Recruiting
    • University Hospital Of Tuebingen
    • Contact: Annika Buchholz, Ph.D.; Phone Number: +4915151819576; Email: annika.buchholz@tuebingen.mpg.de
    • Contact: Juergen Hetzel, M.D.; Phone Number: +491622919339; Email: juergen.hetzel@med.uni-tuebingen.de
    • Sub-Investigator: Bijoy N Atique, M.D.
    • Principal Investigator: Juergen Hetzel, M.D.
    • Sub-Investigator: Maik Haentschel, M.D.

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: N/A
• Genders Eligible for Study: All

• Sampling Method: Probability Sample

Study Population

Patients with detection of SARS-CoV2

Description

Inclusion Criteria:

• Written informed consent
• Age >= 18 years
• Detection of SARS-CoV2 within the past 5 days

Exclusion Criteria:

• Inability to measure vital parameters and document symptoms

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Training cohort; Randomly selection of 80% of the study population. The machine learning algorithm is trained on this dataset	Other: Machine learning; Machine learning on vital parameters, clinical symptoms and underlying diseases
Validation cohort; Randomly selection of 20% of the study population. The machine learning algorithm which was trained on the basis of the training data cohort is validated on the validation cohort.	Other: Machine based evaluation; Quantification of the prediction power and identification of the most relevant predictive parameters

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Probability of Participants for Hospitalisation or Fatal Outcome	Detection of severe acute respiratory syndrome- Corona Virus 2 (SARS-CoV2) to recovery, hospitalisation or fatal outcome up to 5 weeks

Secondary Outcome Measures

Outcome Measure	Time Frame
Probability of Participants for Intensive Care Unit Admission	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Probability of Participants for Fatal Outcome	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Prediction of persisting health impairment by using standardized questionnaires	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Detection of symptoms, vital parameters and comorbidities predicting clinical course	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Influence of size of training data set	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Influence of viral load on the course of disease/ clinical outcome	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Influence of different virus variants on the course of disease/ clinical outcome	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Influence of SARS-CoV2 vaccination (yes/no) on the course of disease/ clinical outcome	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Evaluation of parameters (symptoms, vital parameters, comorbidities) according to their potential of clinical course predictions	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Probability of Participants for hospitalisation	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks
Influence of different SARS-CoV2 vaccines on the course of disease/ clinical outcome	Detection of SARS-CoV2 to recovery, hospitalisation or fatal outcome up to 5 weeks

Collaborators and Investigators

• Sponsor: University Hospital Tuebingen
• Collaborators: Max-Planck-Institute Tuebingen
• Investigators: Study Chair: Bernhard Schoelkopf, PhD, Max-Planck-Institute, Tuebingen, Germany; Principal Investigator: Juergen Hetzel, MD, University Hospital of Tuebingen, Tuebingen, Germany

Study record dates

Study Major Dates

• Study Start (Actual): February 1, 2021
• Primary Completion (Anticipated): July 31, 2021
• Study Completion (Anticipated): December 31, 2021

Study Registration Dates

• First Submitted: March 24, 2021
• First Submitted That Met QC Criteria: March 30, 2021
• First Posted (Actual): April 2, 2021

Study Record Updates

• Last Update Posted (Actual): April 2, 2021
• Last Update Submitted That Met QC Criteria: March 30, 2021
• Last Verified: January 1, 2021

More Information

Terms related to this study

• Keywords: Artificial intelligence; Machine learning; Clinical course
• Additional Relevant MeSH Terms: Pathologic Processes; Coronavirus Infections; Coronaviridae Infections; Nidovirales Infections; RNA Virus Infections; Virus Diseases; Infections; Respiratory Tract Infections; Respiratory Tract Diseases; Pneumonia, Viral; Pneumonia; Lung Diseases; Disease Attributes; Disease Progression; COVID-19; Clinical Deterioration
• Other Study ID Numbers: TEDDI

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No