SBIR Phase I: Artificial intelligence (AI)-enabled ultrasound for imaging and diagnosing musculoskeletal injuries

该小型企业创新研究（SBIR）第一阶段项目的更广泛影响/商业潜力是使超声成像在肌肉骨骼（MSK）损伤中得到广泛采用。在此SBIR期间开发的人工智能（AI）支持的超声图像分析平台可以为更广泛的患者群体带来具有成本效益的诊断成像，因为超声仪器的小尺寸使该技术能够在各种环境中使用，包括医生办公室，物理治疗办公室和体育设施。大约有840，000名临床医生看到MSK损伤的患者。这些临床医生中的每一位都代表着该平台的候选客户，导致美国的总目标市场达到17亿美元。这个小企业创新研究（SBIR）第一阶段项目旨在推进人工智能（AI）支持的超声成像技术。超声检查是目前使用的最依赖于操作者的医学成像方式，限制了超声成像在肌肉骨骼（MSK）评价中的使用。所提出的方法旨在使更广泛地采用超声诊断MSK损伤。该项目涉及开发严格的数学模型，以确定帧序列是否正确对齐，并确定肩部扫描中的肌腱撕裂是否可诊断。将开发一个引导系统，以帮助超声探头放置，从而实现肌腱撕裂存在或不存在的一致识别。所提出的支持AI的超声成像和诊断方法可以帮助从业者定位超声探头，而无需保持精确的角度定位。一旦处于正确的起始位置，医生将根据需要使用引导系统以预先规定的运动移动探头，以进行准确和有效的图像捕获。该技术的目标是使新手从业者能够捕获相关的分析，以帮助提供诊断和治疗计划。该奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to enable widespread adoption of ultrasound imaging for musculoskeletal (MSK) injuries. The ultrasound image analysis platform enabled by artificial intelligence (AI) developed during this SBIR may bring cost-effective diagnostic imaging to a broader patient population as the small form factor of the ultrasound instrument enables the technology to be used in a variety of settings, including doctors’ offices, physical therapy offices, and sports facilities. There are approximately 840,000 clinicians who see patients with MSK injuries. Each of these clinicians represents candidate customers for the platform, leading to a total addressable market in the U.S. of $1.7 billion.This Small Business Innovation Research (SBIR) Phase I project seeks to advance artificial intelligence (AI)-enabled ultrasound imaging technology. Sonography is the most operator-dependent medical imaging modality in use today, limiting the use of ultrasound imaging in musculoskeletal (MSK) evaluation. The proposed approach seeks to enable the wider adoption of ultrasound for diagnosing MSK injuries. This project involves the development of rigorous mathematical models to determine whether a frame sequence is properly aligned and determine if tendon tears in shoulder scans are diagnosable. A guidance system will be developed to aid in ultrasound probe placement so that consistent identification of tendon tear presence or absence is achieved. The proposed AI-enabled approach to ultrasound imaging and diagnostics may help practitioners position the ultrasound probe without the need to maintain precise angular positioning. Once at the correct starting position, the practitioner will move the probe in a pre-prescribed motion using a guidance system as needed for accurate and efficient image capture. The goal of the technology is to enable novice practitioners to capture relevant analyses to assist in providing a diagnosis and a treatment plan.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.