Diagnostic performance of an AI program in detecting Group A Streptococcal Pharyngitis via smartphone-acquired images
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 发起方
- 入组人数
- 200
- 试验地点
- 1
- 主要终点
- Can an AI algorithm detect pharyngitis in a pediatric clinical setting using
研究概览
简要总结
The study will include all 3-11year-old children presenting to the JSS Hospital Emergency Department (ED) and associated departments with complaints of sore throat that fit the inclusion criteria listed above. Principal investigator (PI) or Research Assistant (RA) will recruit patients meeting the inclusion criteria and obtain signed informed consent from the parent/guardian. ED staff will not be required to assist in the obtention of any information. Patients eligible for
participation in the study will be preliminarily identified based on information disclosed during triage upon presentation, followed by full eligibility verification (by the PI or RA) based on information provided by the parent. Recruitment will take place as eligible patients are waiting for physician evaluation. All subjects will be evaluated and screened for study eligibility by the research team prior to study entry. This is a convenience sample of children with pharyngitis; the subjects will be enrolled when a member of the research team is present in the ED. Patient data (name, age, sex, relevant clinical history) will be uploaded into a mobile application by the PI or RA. The identifying information, such as names and dates of birth, of study participants will be omitted from the data library once data collection is complete. The application will consist of a patient input screen, followed by a series of clinical data prompts. Next, a video capture screen where the PI or RA will capture a 3-5 second video of the patients’ throat using the built-in iPhone camera as per recommended steps for video acquisition (see Appendix A.Taking Throat Images with a Smartphone). The treating physician will be blinded to the video take by the PI or RA until they have examined the patient independently. When necessary, a tongue depressor will be used to better visualize the throat.
研究设计
- 研究类型
- Observational
入排标准
- 年龄范围
- 3.00 Year(s) 至 11.00 Year(s)(—)
- 性别
- All
入选标准
- •The study will include all 3-11year-old children presenting to the Montreal Children’s Hospital Emergency Department with complaints of sore throat that fit the inclusion criteria listed above.
- •A Student Investigator (SI) or Research Assistant (RA) (trained in the use of Siurge) will recruit patients meeting the inclusion criteria and obtain signed informed consent from the parent/guardian.
- •ED staff will not be required to assist in the obtention of any information.
- •Patients eligible for.
排除标准
- •Participant parents who are not willing to give consent.
结局指标
主要结局
Can an AI algorithm detect pharyngitis in a pediatric clinical setting using
时间窗: 9 months
images of the oropharynx acquired by a smartphone?
时间窗: 9 months
次要结局
- Outcomes: Sensitivity, specificity, area under the Receiver Operating Characteristic curve([AUROC] and under the precision-recall curve [AUPRC], weighted F1 score, confusion matrices)
研究者
Dr Mandyam Dhati Ravi
JSS HOSPITAL
