Identifying Local Signs at the Catheter Insertion Site With Artificial Intelligence

Catheter Infection Clinical Trial

— DeepCath  

Official title:

Development of an Artificial Intelligence Model of Image Recognition Through Images of Intravascular Catheters From Inpatients and Outpatients to Identify the Presence of Local Signs Associated With Infection

Clinical Trial Details — Status: Recruiting

Administrative data

• NCT number: NCT05440396
• Other study ID #: DeepCath
• Status: Recruiting
• Start date: September 1, 2022
• Est. completion date: December 31, 2023

Study information

• Verified date: October 2022
• Source: Outcome Rea
• Contact: Jean-François TIMSIT, Pr
• Phone: +33140257703
• Email: jean-francois.timsit[@]aphp.fr
• Is FDA regulated: No
• Study type: Observational [Patient Registry]

Clinical Trial Summary

Deepcath is the first step to the introduction of artificial intelligence in catheter care. A better use of visualisation of catheter exit site should be used not only by the HCWs but also by the patients and their family. A deep learning system able to detect visual abnormalities of the catheter exit site will be an helpful tools to develop a continuous follow-up of intravascular catheters.

Recruitment information / eligibility

• Status: Recruiting
• Enrollment: 1000
• Est. completion date: December 31, 2023
• Est. primary completion date: December 31, 2023
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

Patients over 18 years of age Patients with one or more implanted central venous, midline, piCCline, arterial, or peripheral catheters. Patient and/or trusted person and/or family who have verbally stated their non-objection to the study Patient affiliated or beneficiary of a social security plan Exclusion Criteria: Patients presenting a peripheral identification sign close to the catheter insertion point cannot be masked when the photograph is taken. Thus, jewelry, clothing, tattoos, scars, and birthmarks are identifying features. Patients whose catheter insertion point is not visible.

Related Conditions & MeSH terms

• Catheter Infection
• Infections

Intervention

Diagnostic Test:
• Photographs collection phase
Three medical experts have been selected to review the photo collected. Each expert medical assesses the presence of local signs of infection on the photographs by annotating them directly via a dedicated software. They will annotate local signs: redness, perfusion extravasation, necrosis, hematoma, edema, non-purulent discharge, and purulent discharge. A convolutional neural network model will determine the probability of local sign presence. Each picture will be annotated to determine the main characteristics of the catheter. A dataset preparation with photo cropping will be performed for modelling.

Locations

Country	Name	City	State
France	CHU Clermont-Ferrand	Clermont-Ferrand	Puy-de-Dôme
France	CHU Grenoble Alpes	Grenoble	Isère
France	Hôpital Bichat - Claude-Bernard	Paris	Ile De France
France	Réseau RéPIAS SPIADI	Tours

Sponsors (7)

Lead Sponsor	Collaborator
Outcome Rea	Assistance Publique - Hôpitaux de Paris, National Network Surveillance and Prevention of Infections Associated with Invasive Devices SPIADI, UNICANCER, Université Grenoble-Alpes, University Hospital, Clermont-Ferrand, University Hospital, Grenoble

Country where clinical trial is conducted

France, 

References & Publications (6)

Blot S, Ruppé E, Harbarth S, Asehnoune K, Poulakou G, Luyt CE, Rello J, Klompas M, Depuydt P, Eckmann C, Martin-Loeches I, Povoa P, Bouadma L, Timsit JF, Zahar JR. Healthcare-associated infections in adult intensive care unit patients: Changes in epidemio — View Citation

Buetti N, Abbas M, Pittet D, Chraiti MN, Sauvan V, De Kraker MEA, Boisson M, Teixeira D, Zingg W, Harbarth S. Lower risk of peripheral venous catheter-related bloodstream infection by hand insertion. Antimicrob Resist Infect Control. 2022 Jun 3;11(1):80. — View Citation

Buetti N, Mermel LA, Timsit JF. Routine catheter-tip cultures for assessing catheter-related bloodstream infections in randomised-controlled trials. Anaesth Crit Care Pain Med. 2022 Feb;41(1):101006. doi: 10.1016/j.accpm.2021.101006. Epub 2021 Dec 14. — View Citation

Buetti N, Rickard CM, Timsit JF. Catheter dressings. Intensive Care Med. 2022 Aug;48(8):1066-1068. doi: 10.1007/s00134-022-06734-w. Epub 2022 May 27. — View Citation

Timsit JF, Schwebel C, Bouadma L, Geffroy A, Garrouste-Orgeas M, Pease S, Herault MC, Haouache H, Calvino-Gunther S, Gestin B, Armand-Lefevre L, Leflon V, Chaplain C, Benali A, Francais A, Adrie C, Zahar JR, Thuong M, Arrault X, Croize J, Lucet JC; Dressi — View Citation

Timsit JF. Diagnosis and prevention of catheter-related infections. Curr Opin Crit Care. 2007 Oct;13(5):563-71. Review. — View Citation

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	The number of correct predictions for redness divided by the total number of predictions	Overall classification accuracy of the learning model compared to the assessment of three independent medical experts on the detection of the presence of redness greater than or equal to 5 mm at the catheter insertion site.	through study completion, an average of 1 year
Secondary	The ratio of true positives and total positives predicted:	The precision metric focuses on Type-I errors(FP). A Type-I error occurs when we reject a true null Hypothesis.	through study completion, an average of 1 year
Secondary	The number of correct predictions for indurated venous cord divided by the total number of predictions	Evaluate, on the basis of images of peripheral catheter insertion sites, the reliability of the learning model on the assessment of the presence of indurated venous cord.	through study completion, an average of 1 year
Secondary	The link between presence of local signs and infection	Measure the correlation between the appearance of the catheter puncture site and the presence of signs consistent with local and systemic infection.	through study completion, an average of 1 year