Clinical Decision Support and Medical Image Analysis using Artificial Intelligence

使用人工智能的临床决策支持和医学图像分析

• 批准号: 2457898
• 金额: --
• 依托单位: University of Kent
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2457898
• 关键词: Clinical; Decision; Support; Medical; Image

"The aim of the project is to develop AI that will perform clinical decision support for the purpose of assisting clinicians in writing individual patient treatment plans. Deep neural networks will be trained on diverse sets of medical data to learn complex relationships between symptoms and treatments. An approach to this problem could be:1) treatment planning using knowledge graphs and intelligent reasoning,2) medical image analysis by convolutional neural networks (CNN), and3) medical record parsing using natural language processing (NLP) frameworks, such as the BERT framework and the Transformer network.The successful applicant will be jointly supervised by Dr Gibson and Clinicians within the East Kent Hospitals University NHS Foundation Trust and will benefit from existing collaborations with Ophthalmology and Urology at Kent and Canterbury Hospital. Protocols and ethical approval for accessing training and evaluation data are already in place."

“该项目的目的是开发人工智能，该人工智能将执行临床决策支持，以协助临床医生编写个体患者治疗计划。深度神经网络将接受不同医疗数据集的训练，以了解症状和治疗之间的复杂关系。解决这个问题的方法可能是：1）使用知识图和智能推理进行治疗计划，2）通过卷积神经网络（CNN）进行医学图像分析，3）使用自然语言处理（NLP）框架（例如 BERT）进行医疗记录解析框架和Transformer 网络。成功的申请人将由吉布森博士和东肯特大学 NHS 基金会信托基金内的临床医生共同监督，并将受益于与肯特医院和坎特伯雷医院眼科和泌尿科的现有合作以及获取培训和评估数据的伦理批准。已经就位了。”