SBIR Phase I: A Powerful Clinical Aid in the Diagnosis of Vaginitis to Prevent PTB and Stop STI Transmission

这个小型企业创新研究(SBIR)第一阶段项目的更广泛的影响/商业潜力是提供快速、准确的阴道感染诊断。阴道炎每年导致1000万次就诊，占美国所有女性健康就诊的10%，众所周知，阴道炎极大地增加了性传播感染的易感性。外部检测费用高昂，可能需要一天到几周的时间才能得出结果。在目前的办公室检测中，40%的酵母菌感染和30%的细菌感染被误诊，需要进一步检查。更重要的是，阴道炎可能导致很大一部分自发早产--这是全球新生儿死亡、情绪困扰和终身残疾的主要原因。在美国，肺结核每年导致260亿美元的医疗成本，这促使人们进行快速而准确的检测。该项目将开发一种高级测试，使用人工智能(AI)在医生办公室快速准确地研究样本。这个小企业创新研究第一阶段项目推进了一种针对临床表现的阴道炎的新型测试。该项目利用人工智能(AI)图像分析和高质量的荧光显微镜来快速扫描医生办公室的标本，以进行快速诊断。该项目将验证初步研究，这些研究表明，与在显微镜下观察类似样本相比，图像精度更高。该项目的第一阶段将检查从多达500名妇女的不同患者群体中准备好的样本，以优化VHA自动数字病理系统的分析和表征。所有样本都将与目前临床使用的人工评估进行比较。此外，将在多种类型的临床研究环境中进行用户测试，如医学院和健康网络，以使患者样本群体多样化。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 Small Business Innovation Research (SBIR) Phase I project is to provide fast, accurate diagnosis of vaginal infections. Vaginitis results in 10 million office visits a year representing 10% of all US women’s health visits, is known to dramatically increase susceptibility to sexually transmitted infections. External testing is expensive and can take from a day to weeks for results. For current in-office tests, 40% of all yeast infections and 30% of bacterial infections are misdiagnosed and require further visits. More importantly, vaginitis can lead to a large fraction of spontaneous preterm births - a major cause of neonatal mortality, emotional distress and lifelong disability worldwide. In the US, PTB results in healthcare costs of $26 B/year, motivating a fast and accurate test. This project will develop an advanced test that uses artificial intelligence (AI) to study samples quickly and accurately in the doctor's office.This Small Business Innovation Research Phase I project advances a novel test for vaginitis upon clinical presentation. This project leverages Artificial Intelligence (AI) image analysis with a high-quality fluorescence microscope to rapidly scan specimens in the doctor's office for fast diagnosis. This project will validate preliminary studies suggesting higher image accuracy compared to viewing similar samples under a microscope. Phase 1 of this project will examine prepared samples from diverse patient populations of up to 500 women to optimize analysis and characterization of VHA automated digital pathology system. All samples will be compared to manual assessment currently used clinically. Additionally, user testing in multiple types of clinical research environments, such as medical schools and health networks, will be undertaken to diversify patient sample populations.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.