SBIR Phase II: Acoustoelastic Tissue Property Evaluation of Selected Tissue Region in Dynamic Ultrasound Images

这个小型企业创新研究(SBIR)第二阶段项目计划通过将第一阶段开发的新型超声后处理软件应用到可编程平台超声系统中，开发用于评估肌肉骨骼软组织状况的实时超声系统。今天，放射科医生通过观察静态核磁共振或常规超声图像，并考虑仅支持定性、主观评估的关键因素来诊断大多数肌肉骨骼疾病。开发一种高效、实时、定量的方法来诊断软组织(如肌腱和韧带)损伤并监测愈合情况，可以导致更准确的诊断，并减少未完全愈合的组织的再损伤。该项目将提高新软件技术的临床实用性和工作流程效率。原有的软件将通过改进软件来增强，以自动检测超声图像中的感兴趣区域。登记的感兴趣区域可以从一次病人就诊到下一次病人来访精确匹配。通过利用机器学习来辅助诊断决策，开发数据挖掘软件将进一步提高效率和准确性。这些软件改进将与平台超声系统集成，以改善临床工作流程。集成的产品将与标准超声波的工作流程效率相匹配，并极大地提高超声波在肌肉骨骼领域的应用。如果该项目获得成功，其更广泛的影响/商业潜力将大大提高临床医生护理软组织损伤的能力，并使该公司能够利用(1)降低医学成像成本的压力，(2)肌肉骨骼专家对超声波，特别是便携式仪器的日益浓厚的兴趣，(3)一家主要的超声波制造商专注于规模较大且相对尚未开发的肌肉骨骼超声市场，以及(4)最近出现的用于非肌肉骨骼应用的定量超声波。这个第二阶段的项目将产生一种有效的、实时的、定量的诊断软组织损伤和监测愈合的方法。仅在美国，过度使用损伤(拉伤、扭伤)是最常见的肌肉骨骼损伤。每年，1840万这样的伤害花费了大约9200万美元。肌肉骨骼损伤的患者目前在诊断、护理和结果方面面临三个挑战。首先，目前的诊断方法，包括MRI、超声波或物理操作，依赖于高度主观和依赖于观察者的解释，因此准确性各不相同。其次，核磁共振仍然是治疗的标准，但比超声波贵得多。在适当的情况下，用超声波代替核磁共振对肌肉骨骼疾病进行初步诊断，每年可以为医疗保险节省7.36亿美元。第三，缺乏客观的监测方法来确定患者何时可以安全地恢复活动，这意味着患者有可能错过不必要的工作时间或再次受伤。

This Small Business Innovation Research (SBIR) Phase II project proposes to develop a real-time ultrasound system for evaluating musculoskeletal soft tissue conditions by implementing the novel ultrasound post-processing software developed in Phase I into a programmable platform ultrasound system. Today, radiologists diagnose most musculoskeletal diseases by observing static MRI or conventional ultrasound images and considering key factors that support only qualitative, subjective assessments. Developing an efficient, real-time, quantitative method for diagnosing soft tissue (e.g., tendons and ligaments) injuries and monitoring healing can lead to more accurate diagnoses and reduce re-injury of incompletely healed tissues. The project will enhance the novel software technology's clinical utility and workflow efficiency. The original software will be enhanced by improving the software to automatically detect a region of interest with the ultrasound image. The registered regions of interest can be matched precisely from one patient visit to the next. Developing data mining software will further increase efficiency and accuracy by leveraging machine learning to assist with diagnostic decisions. These software improvements will be integrated with the platform ultrasound system to improve clinical workflow. The integrated product will both match the work flow efficiency of standard ultrasound and dramatically advance the utility of ultrasound within the musculoskeletal arena. The broader impact/commercial potential of this project, if successful, will dramatically improve clinicians' ability to care for soft tissue injuries and will position the company to capitalize on (1) pressure to reduce medical imaging costs, (2) musculoskeletal specialists' growing interest in ultrasound, especially portable instruments, (3) a major ultrasound manufacturer's focus on the large and relatively untapped musculoskeletal ultrasound market, and (4) the recent emergence of quantitative ultrasound for non-musculoskeletal applications. This Phase II project will produce an efficient, real-time, quantitative method for diagnosing soft tissue injuries and monitoring healing. In the US alone, overuse injuries (strains, sprains) are the most frequently reported musculoskeletal injuries. Annually, 18.4 million such injuries cost approximately $92 B. Patients suffering from musculoskeletal injuries currently face three challenges at diagnosis, care, and outcome. First, current diagnostic methods, including MRI, ultrasound, or physical manipulation, rely on highly subjective and observer-dependent interpretation, so accuracy varies. Second, MRI is still the standard of care, but is far more costly than ultrasound. Substituting ultrasound for MRI, where appropriate for initially diagnosing musculoskeletal conditions, could save Medicare $736 M/year. Third, the lack of an objective monitoring method to determine when a patient can safely return to activity means patients risk missing unnecessary work time or re-injury.