Explainable AI recommendations for infectious disease management: providing clear, robust, and reliable communication to users

针对传染病管理的可解释人工智能建议：为用户提供清晰、稳健和可靠的沟通

• 批准号: 10073509
• 金额: $ 6.37万
• 依托单位: NEX.Q LTD
• 依托单位国家: 英国
• 项目类别: Grant for R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10073509
• 关键词: Explainable; AI; recommendations; infectious; disease

Effective decision-making is crucial to control outbreaks of infectious diseases. Poor decision-making around infection prevention and control costs the NHS an estimated £2.7 billion annually in the UK. This is primarily due to insufficient time and resources for teams to properly analyse outbreak situations and determine the most effective actions to take. These inefficiencies lead to prolonged lengths of stays, unnecessary medical procedures, and severe consequences for patients.Infection prevention and control is a complex field, and artificial intelligence (AI) has the potential to revolutionise how infectious diseases are managed by supporting optimal decision-making. However, the black-box nature of these AI systems and the inability to communicate effectively has hindered their adoption at scale. As a result, we are applying for funding to develop trust, transparency, and communication into a novel AI recommendation tool for infection prevention and control teams. By providing this level of transparency and communication, we hope to increase trust in AI tools and promote their adoption in infection prevention and control.The potential impacts of this system are far-reaching, not only in the primary marketplace of hospital infection prevention and control but also in providing transparency and communication of algorithm recommendation systems, enabling them to be adopted in various use cases. With the development of this system, we hope to decrease the costs associated with poor decision-making in infection prevention and control and ultimately improve patient outcomes. Improving the efficiency of decision-making in this field can help reduce the spread of infectious diseases and prevent future outbreaks.Investing in developing such systems is critical to provide better care and reducing healthcare costs. We believe that our research at NEX.Q will lead to the development of a valuable tool that can be adopted in various healthcare settings and ultimately improve patients' lives.

有效的决策对于控制传染病的爆发至关重要。在英国，围绕感染预防和控制的糟糕决策每年给 NHS 造成约 27 亿英镑的损失，这主要是因为团队没有足够的时间和资源来正确分析疫情。这些低效率会导致住院时间延长、不必要的医疗程序以及给患者带来严重后果。感染预防和控制是一个复杂的领域，人工智能 (AI) 有潜力彻底改变感染预防和控制的方式。传染病是然而，这些人工智能系统的黑匣子性质和无法有效沟通阻碍了它们的大规模采用。因此，我们正在申请资金来发展信任、透明度和沟通。为感染预防和控制团队提供一种新颖的人工智能推荐工具，通过提供这种程度的透明度和沟通，我们希望增加对人工智能工具的信任，并促进其在感染预防和控制中的采用。该系统的潜在影响是深远的。不仅在医院感染防控的一级市场，而且在提供算法推荐系统的透明度和沟通性，使其能够在各种用例中采用，我们希望减少与感染预防和控制方面的不良决策相关的成本，并最终改善患者的治疗效果。该领域决策的效率有助于减少传染病的传播并预防未来的爆发。投资开发此类系统对于提供更好的护理和降低医疗保健成本至关重要。我们相信，我们在 NEX.Q 的研究将带来以下成果。开发可用于各种医疗保健的有价值的工具环境并最终改善患者的生活。