SBIR Phase I: Aneurisk - A Clinical Decision Support Tool to Manage Abdominal Aortic Aneurysm Patients

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project will provide a novel software solution that enables clinicians to potentially improve the treatment of patients with an abdominal aortic aneurysm (AAA). Abdominal aortic aneurysm is the ballooning of a major blood vessel in the body that if left untreated can rupture, leading to almost certain (85% mortality) death. AAA is the 13th leading cause of death (1 in 100,000) in the United States. The current clinical standard for surgical intervention relies on measuring the diameter of the aneurysm from medical images. This undesirable method leads up to 23.4% of patients rupturing before reaching the diameter that indicates safe, surgical intervention/repair. Therefore, there is a need to help doctors identify and treat high-risk patients who remain below the threshold of safe aneurysm diameter. This STTR project supports the development of a unique, artificial intelligence, solution that combines aneurysm diameter with stress, strain, shape analysis, and patient information to predict future patient outcomes and aneurysm growth. The patent-pending technology can disrupt how doctors currently watch and follow aneurysms by providing them with a risk profile to avoid dangerous rupture events allowing them to provide the patients with the right treatment at the right time This Small Business Innovation Research (SBIR) Phase I project aims to develop and validate machine learning models for risk classification, growth projection, and wall stress prediction for abdominal aortic aneurysms. Abdominal aortic aneurysm (AAA) is the 13th leading cause of death in the United States, with a mortality rate exceeding 85%. The current clinical standard for intervention is based on the diameter of the aneurysm, however, between 7 and 23.4% of patients rupture before the threshold is reached. The Aneurisk team is developing artificial intelligence-based tools to accelerate image, biomechanical, and morphological analyses. Additionally, the Aneurisk team will perform cross-validation of previously trained long short-term memory recurrent neural networks to forecast diameter and a classifier that predicts patient outcomes (remain stable, eventual repair, or eventual rupture). The toolset is effectively a method to achieve virtual surveillance to provide clinicians a method to better understand when to treat patients. If successful, the Aneurisk approach will reduce the number of surveillance visits to track diameter, reduce patient anxiety, and reduce costly and high-mortality rupture events. 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.