Assessment of an App Based on Artificial Intelligence for Purulent AOM Diagnosis in a Pediatric Department

Not Applicable

Completed

• Conditions: Otitis Media, Suppurative
• Interventions: Other: AOM diagnosis with app; Other: AOM diagnosis without app

• Registration Number: NCT04752891
• Lead Sponsor: Fondation Lenval

Brief Summary

Otitis media (OM) is one of the most common childhood infections and is a major cause of morbidity in children and results as being the first cause of antibiotic prescription among children in developed countries . An artificial intelligence-based tool could help physicians refine their diagnosis

Detailed Description

Ear infections in children are very common, particularly acute otitis media (AOM) which represents the first reason for visiting physicians and for antibiotic prescription in children in developed countries. However, the diagnosis is not easy because of the narrowness of the ear canal, the discomfort of the child during the otoscopic exam. Moreover, the accuracy of the AOM diagnosis depends on the subjectivity of the physician mostly. This can result in an misdiagnosis and inappropriate treatment with, as a result, the emergence of multi-resistant germ.

Artificial intelligence (AI) in medicine is booming and has already proven its worth in terms of prevention, monitoring and diagnosis. I-Nside, an app based on Artificial Intelligence is able to make automatic diagnosis of many ear pathologies, after capturing picture of any eardrum using an otoendoscope fixed on a connected smartphone (SmartScope®).

Therefore, the investigators aim to assess the accuracy of AOM diagnosis the existing app, by comparing the inter-rater agreement between the diagnosis made by a senior physician using a traditional otoscope (considered as "gold standard" ) and the one made by a junior physician using the app

Study Timeline

• Study First Submitted: 2021-02-02
• Study First Submitted QC: 2021-02-11
• Study First Posted: 2021-02-12
• Study Start: 2021-06-10
• Primary Completion: 2022-08-16
• Status Verified: 2023-03
• Study Completion: 2023-03-16
• Last Update Submitted: 2023-03-16
• Last Update Posted: 2023-03-17

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 203

Inclusion Criteria

Child under 18 years Presenting signs that may suggest acute otitis media (fever, earache, purulent otorrhea) For which the informed written consent of, at least, one of the two parents or the holder of parental authority has been obtained

Exclusion Criteria

A sign (s) of vital distress Trans-tympanic drains

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SEQUENTIAL

Arm && Interventions

Group	Intervention	Description
AOM diagnosis with app	AOM diagnosis with app	-
AOM diagnosis without app	AOM diagnosis without app	-

Primary Outcome Measures

Name	Time	Method
diagnosis of purulent Acute Otitis Media (AOM) made by junior physicians using the i-Nside application associated with Smartscope®	at inclusion	diagnosis of purulent AOM is made on an inflammatory aspect of the eardrum with retrotympanic effusion, exteriorized (otorrhea) or not (opacity, effacement of reliefs or bulging, disappearance of the light triangle).
diagnosis of purulent Acute Otitis Media (AOM) made by senior physicians using a standard otoscope	at inclusion	diagnosis of purulent AOM is made on an inflammatory aspect of the eardrum with retrotympanic effusion, exteriorized (otorrhea) or not (opacity, effacement of reliefs or bulging, disappearance of the light triangle)

Secondary Outcome Measures

Name	Time	Method
rate of misdiagnoses of purulent Acute Otitis Media (AOM) with and without the use of the application	at inclusion	a misdiagnosis is measured by a discordant diagnosis of the juniors compared to the seniors considered as the Gold Standard for the AOM diagnosis
satisfaction of junior physicians regarding the use of the app	through study completion, an average of 1 year	a questionnaire will be given at the end of the study. The questions relate to the use of the application and its perceived usefulness by practitioners. Satisfaction will be assessed by a Lickert scale
collection risk factors for a discrepancy in the diagnosis of purulent AOM	at inclusion	collection data of age of patients,pain requiring analgesia, period of the day of diagnosis (morning, afternoon,night)

Trial Locations

Locations (1)

Hôpitaux Pédiatrique de Nice CHU Lenval; 🇫🇷 Nice, France