I-Corps: A Clinical Decision Support Tool to Manage Abdominal Aortic Aneurysm Patients

I-Corps项目更广泛的影响/商业潜力是开发一种依赖于机器学习最新进展的方法，机器学习在患者疾病的诊断和预后方面产生了积极的结果。美国有670万脑动脉瘤患者和250万腹主动脉瘤患者。患有动脉瘤的患者被比作拥有一枚定时炸弹，可以在任何时候引爆，由于内出血过多而导致死亡或发病。来自医学图像的生物力学和形状分析可以更好地了解动脉瘤是如何随着时间的推移而发展的。然而，医学图像的分析是临床工作流程的一个重大瓶颈，减少了将此类工具转化为临床环境的工作。拟议的技术可以减少从医学图像中提取关键信息的总体时间，从而减少对计算专家提供基于人工智能的方法来识别患者风险的需求。从各种指数个性化药物的能力允许临床医生从压力和形状分析中导入信息或数据点，这些信息和数据点通常是非专家无法访问的。这个i-Corps项目基于动脉瘤预后分类器的开发，该分类器将针对小动脉瘤进行培训，为临床医生决定昂贵的手术干预措施创造一个强大的风险评分。该方法将跟踪变化，从单个患者在不同时间点的不同扫描中确定动脉瘤的形状和生物力学指数，这将有助于确定动脉瘤随时间的演变。临床医生将能够根据各种因素决定手术干预风险，而不是依赖最大直径标准，并将能够通过缩短监测间隔(从而减少监测的总体成本)来个性化患者保健管理。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is the development of a methodology that relies on recent advancements in machine learning, which have yielded positive results in diagnosis and prognosis of diseases in patients. There are 6.7 million cerebral aneurysm patients and 2.5 million abdominal aortic aneurysm patients in the United States. Patients that suffer from aneurysms are likened to having a ticking timebomb that can set off at any second, causing death or morbidity due to excessive internal bleeding. Biomechanical and shape analyses from medical images can provide greater insight into how aneurysms progress with time. However, analyses of medical images are a significant bottleneck to a clinical workflow, reducing the translation of such tools into a clinical setting. The proposed technology can reduce the overall time to extract critical information from medical images, alleviating the need for computational experts to provide an artificial intelligence-based approach to striate patient risks. The ability to personalize medicine from various indices allows clinicians to import information or data points from stress and shape analyses that are typically inaccessible to non-experts.This I-Corps project is based on the development of an aneurysm prognosis classifier that will be trained on small aneurysms, creating a robust risk score for clinicians to decide on costly surgical interventions. The approach will track changes to determine the shape of the aneurysm and biomechanical indices from various scans for a single patient at different time points, which will aid in determining the evolution of the aneurysm with time. A clinician would be able to decide on surgical intervention risk based on various factors rather than relying on the maximum diameter criterion and would be able to personalize patient healthcare management by reducing the surveillance intervals (whereby reducing overall costs of monitoring).This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.