A Volunteer Study to Collect Imaging Data for the Development of the IntelligentUltrasound Anatomy Guide
试验速览
- 阶段
- 不适用
- 状态
- 已完成
- 发起方
- 入组人数
- 150
- 试验地点
- 1
- 主要终点
- Model development and verification
研究概览
简要总结
This is a single-centre, prospective, non-randomised volunteer study to be undertaken in the MediCentre, Heath Park, Cardiff, UK
详细描述
This is a single-centre, non-randomised, prospective study involving at least 120 participants. The data collected in this study will augment the data collected in IU2019_AG_03 Volunteers will be enrolled sequentially until at least 120 scans have been recorded. Ideally, between 120 and 150 scans will be recorded.
Phase I Data collected from up to the first 80 participants in each of the six peripheral nerve block areas will be used as a training/verification set. This set will be used to develop and verify models for the identification and highlighting of target anatomical structures on ultrasound images.
Phase II The remaining 40 videos in each area (total minimum 120 participants) will be used for validation of these models.
The validation videos will be fed into the models created during Phase I of the study and the output anatomy highlighting will be recorded as a separate video, overlaid on the original ultrasound image. The overlay video will then be further split into 30-second segments.
A panel of three independent expert anaesthetists will be convened to review and score the segments for accuracy and performance to the endpoints agreed in Phase I of the study.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 是
入选标准
- •Male or female, at least 18 years of age;
- •Able to comprehend and sign the Informed Consent prior to enrolment in the study.
排除标准
- •Aged <18 years of age;
- •Unwilling or unable to provide informed consent.
结局指标
主要结局
Model development and verification
时间窗: 4 months
Collection of ultrasound scans. The collected scans will be assessed and used in the development and verification of deep-learning models that identify the target structures using a training dataset.
次要结局
- Model validation(4 months)