Artificial Intelligence in Modelling the Influence of Socio-Economic Factors on the Risk of Cardiovascular Events

学生战略优先领域：数学、统计和计算关键词：心血管疾病、人工智能心血管疾病(CVDS)是全球发病率和死亡率的主要原因，特别是在老年人中，尽管诊断和治疗取得了实质性进展。人工智能(AI)的应用提供了最先进的数据建模和解释，以提供信息和支持临床决策。机器学习和深度学习等人工智能技术可以根据纵向健康记录提供对医疗选择的高级见解，从而提高医学知识。医疗保健和人工智能这一新兴的多学科领域在向人们和医疗保健专业人员告知医疗事件和决定如何与结果相关并可能影响结果方面将变得越来越重要。然而，医疗事件可能会受到富裕和社会贫困的强烈影响。我的研究将使用来自Greater Glasgow&Clyde Health Board的NHS管理数据，用于50岁的成年人(影响老年人的疾病发病率和流行率迅速增加的年龄)。这包括一个庞大的数据集，包括人口统计学、血液测试、心电图和超声心动图、初级保健处方、住院和程序以及死亡率。对于较大的子集，可以获得关于吸烟、血压和体重指数的额外初级保健数据。我将使用这些数据来探索常见的事件顺序(例如，吸烟、高血压和肥胖，导致糖尿病、肾功能障碍和动脉粥样硬化，并继而导致心肌梗死、心力衰竭、中风、残疾和死亡)以及考虑年龄、它们与富裕和社会剥夺的关系。

Studentship strategic priority area: Mathematics, Statistics and ComputationKeywords: Cardiovascular Disease, Artificial Intelligence Cardiovascular diseases (CVDs) are leading cause of morbidity and mortality worldwide, especially in older people, despite substantial advances in diagnosis and treatment. Application of artificial intelligence (AI) offers state-of-the-art data modelling and interpretation to inform and support clinical decisions. AI techniques such as machine learning and deep learning can improve medical knowledge, by providing advanced insights into healthcare choices based on longitudinal health records. The emerging multidisciplinary field of healthcare and AI will become increasingly important in informing people and healthcare professionals about how medical events and decisions are associated with and potential influence outcome. However, medical events may be strongly influenced by affluence and social deprivation. My research will use NHS administrative data from Greater Glasgow & Clyde Health Board for adults age >50 years (the age at which the incidence and prevalence of disease affecting older people increases rapidly). This comprises a large dataset including demographics, blood tests, electrocardiograms and echocardiograms, primary-care prescriptions, hospitalisations and procedures and mortality. For a large subset, additional primary care data on smoking, blood pressure and body mass index can be obtained. I will use these data to explore the common sequence of events (e.g. smoking, hypertension and obesity, leading to diabetes, renal dysfunction and atherosclerosis and onwards to myocardial infarction, heart failure, stroke, disability and death) and accounting for age, their relationship to affluence and social deprivation.