iToBoS Prospective Data Acquisition Study
试验速览
- 阶段
- 不适用
- 状态
- 招募中
- 入组人数
- 350
- 试验地点
- 1
- 主要终点
- Incidence of skin diagnostic categories
研究概览
简要总结
This study is recruiting patients at their regular skin examination appointments to participate in research. Participation involves having 3D total body photography, completing a 10-15 minute questionnaire, and providing a genetic sample. Normally, the total body photography is part of the patients standard care, as is the collection of a genetic sample. Consenting to this study involves consenting to the use of total body photography images (de-identified), questionnaire answers, and genetic risk information to be used for developing AI algorithms for image analysis of skin lesions, and melanoma-risk profiling for patients.
详细描述
Patients are invited to participate at their regular skin examination visits at Hospital Clinic Barcelona. A member of the research team will read through the Participant Information and Consent Form (PICF) with the patient and answer any questions they may have. Once consent is given, the following study activities will take place.
- Total Body Photography: this is done with the VECTRA machine that will require you to be photographed in your underwear. The VECTRA is a framework of 92 cameras which will simultaneously capture images of a patient holding one anatomical pose. VECTRA computer software is then used to construct a 3D avatar of the patient, enabling the image record of skin naevi. This imaging is normally part of the patient's skin examination visit. The images used for annotation in research will be de-identified (anonynmised) by firstly removing any images of the face and other identifying features (e.g. tattoos, scars, nipples), then processing the 3D avatar into small 2D images of 5x8cm section of skin. These methods are designed to protect patient privacy of images used for AI development.
- Complete a Questionnaire: consists of 45 items/questions, taking approximately 10 minutes to complete. Questionnaire data is designed to collect information relevant to melanoma risk, while maintaining a low risk of being individually identifiable. topics cover include demographics, skin cancer history, sun/health behaviour, and phenotype information (e.g. hair/skin/eye colour).
- Genetic sample: in some cases the hospital will already hold a genetic sample, in this case we ask patient consent to access this information. When no sample is on record, we will ask the patient to provide a saliva or blood sample for genetic analysis. Genetic information will be analysed by looking for single nucleotide polymorphisms (SNPs) relating to melanoma risk (approximately 80-100 SNPs to be analysed). A polygenic risk score (PRS) is then calculated based on the presence or absence of relevant SNPs. The final PRS will be used in the study (no individual genetic information will be shared outside the hospital.
All de-identified data (image, survey, and PRS) are then used by the iToBoS consortium to develop a cognitive assistant (AI model) to provide clinician support in screening for melanoma using total body photography.
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
- •Able to provide informed consent
排除标准
- •Unable to stand and hold anatomica pose for 3D total body photography
结局指标
主要结局
Incidence of skin diagnostic categories
时间窗: Through study completion, an average of 1 year
Images will be used to label different skin diagnostic categories. These labelled images are then used as a training dataset for machine learning in developing algorithms for skin lesion diagnosis.
次要结局
未报告次要终点
研究者
Clare Primiero
Associate Investigator
The University of Queensland
