mAIcare: AI for Self-Management of Chronic Illness

点击翻译按钮获取中文摘要

Long-term conditions account for 70% of total medical spending in the UK, and up to 90% in the US. The costs are set to rise as the population is ageing rapidly. Unless there is a change in how long-term conditions are managed, healthcare will become less accessible and more expensive. A solution to this may be self-management, where patients or their carers recognise symptoms and learn how to act on them promptly. We will develop an innovative approach that uses intelligent algorithms to help patients to self-manage better. Initially, we will focus on Chronic Obstructive Pulmonary Disease (COPD) - a progressive lung disease and the third leading cause of death worldwide. In many countries, COPD has already surpassed chronic heart disease as the main cause of hospital admissions. About 30% of patients discharged after a hospital admission for COPD die within 180 days, and about 50% die within two years. COPD kills more people than the second deadliest cancer, and it kills as many UK women as the deadliest cancer. And while death rates from other major causes are declining, COPD is on the rise. Warning signs and self-management plans that work for one person with COPD may not work for others, and it is now widely recognized that self-management of COPD needs to be an ongoing process individualised to each patient in order to be successful. Unfortunately, primary care providers are often overworked and have insufficient time and resources to provide tailored continual support, which reduces self-management programmes to one-off one-size-fits-all "things" given to patients. Our project will develop artificial intelligence (AI) that will adapt to patients and help them to self-manage better. Patients will be able to receive algorithmic feedback on their self-management techniques 24/7, without having to go to hospital or wait for appointments. Healthcare professionals will be able to make adjustments to self-management plans if they receive warnings from algorithms. A part of the project will help COPD patients to cope with anxiety. Our approach will put citizens in control of the illness, and encourage them to engage in evidence-based risk avoidance strategies approved by clinicians. This may both improve patient outcomes and make healthcare more sustainable. If our approach is successful, it can be extended to many other long-term diseases.