Late stage development of a skin cancer detection service

皮肤癌检测服务的后期开发

• 批准号: 105286
• 金额: $ 95.58万
• 依托单位: SKIN ANALYTICS LTD
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=105286
• 关键词: Late; stage; development; skin; cancer

"Skin Analytics has developed an Artificial Intelligence (AI) solution for the detection of skin cancer from images taken by a smartphone camera with a dermoscopic lens attachment. Our vision is to reduce the number of deaths from the disease while significantly reducing the cost of finding skin cancer.We have shown our skin cancer detection service (called DERM) can identify cases of malignant melanoma (MM), Basal Cell Carcinoma (BCC) and Squamous Cell Carcinoma (SCC) with a high degree of accuracy from historical images of skin lesions. We have conducted a clinical validation study with seven NHS Trusts on the melanoma detection algorithm, and in this project we will do the same for the non-melanoma skin cancers (NMSC). We will also conduct research into whether we can detect changes in lesions over time, which would be important for doctors who want to 'watch and wait' suspicious lesions. This project includes the clinical studies required to complete this work.A large part of the project aims to integrate DERM into the NHS. This involves both the technical integration with the systems used by GPs etc, and securing the relevant regulatory clearances, but also gaining an understanding of the patient pathways and how DERM might affect them, exploring barriers to adoption, and gaining user feedback on the soaftware. We will also conduct an initial health economic analysis of the costs/benefits of a skin cancer service in the NHS.At the end of the project Skin Analytics will have sufficient clinical, user experience, health economic data and regulatory clearance to support the implementation of DERM in different clinical practices across the NHS."

“Skin Analytics开发了一种人工智能（AI）解决方案，用于从带有皮肤镜透镜附件的智能手机摄像头拍摄的图像中检测皮肤癌。我们的愿景是减少皮肤癌的死亡人数，同时大幅降低皮肤癌的诊断成本。我们的皮肤癌检测服务（称为DERM）可以从皮肤病变的历史图像中高度准确地识别恶性黑色素瘤（MM）、基底细胞癌（BCC）和鳞状细胞癌（SCC）。我们已经与七家NHS Trust对黑色素瘤检测算法进行了临床验证研究，在本项目中，我们将对非黑色素瘤皮肤癌（NMSC）进行同样的研究。我们还将研究我们是否可以检测到病变随时间的变化，这对于那些想要“观察和等待”可疑病变的医生来说非常重要。该项目包括完成这项工作所需的临床研究。该项目的很大一部分旨在将DERM纳入NHS。这涉及与全科医生等使用的系统的技术集成，并确保相关的监管许可，但也要了解患者的途径以及DERM如何影响他们，探索采用的障碍，并获得用户对软件的反馈。我们还将对NHS中皮肤癌服务的成本/效益进行初步的健康经济分析。在项目结束时，皮肤分析将拥有足够的临床，用户体验，健康经济数据和监管许可，以支持在NHS的不同临床实践中实施DERM。"