STTR Phase I: Using AI to develop a red blood cell health index for the monitoring of sickle cell disease

STTR 第一阶段：利用人工智能开发红细胞健康指数，用于监测镰状细胞病

• 批准号: 2112027
• 负责人: Timothy Adamson
• 金额: $ 25.59万
• 依托单位: KOVADX INC
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2112027&HistoricalAwards=false
• 关键词: STTR; Phase; Using; AI; develop

The broader impact of this Small Business Technology Transfer (STTR) Phase I project is to reach underserved communities and address inequities in health care for patients with hemolytic anemias by providing fast, affordable, and accurate diagnosis and monitoring of hemolytic anemias by combining 3D phase imaging with deep learning. Sickle Cell Disease (“SCD”) is a global health problem that significantly impacts the life span, quality of life, and health outcomes of affected individuals. SCD is one of the most common hemolytic diseases in Sub-Saharan Africa and the U.S, affecting up to 3% of the newborn population. However, few resources and research are dedicated to improving the diagnosis and monitoring of SCD. Individuals who lack access to screening and testing are susceptible to an early-life mortality rate of up to 90%. Unfortunately, those who are most likely to suffer from hemolytic diseases like SCD are frequently underserved by advanced health care. This project may significantly reduce the SCD burden by providing monitoring to prevent crises that require hospitalizations and emergency care. This Small Business Technology Transfer (STTR) Phase I project will advance Artificial Intelligence (AI) innovations for diagnosing red blood cell disease with quantitative phase imaging (QPI). While QPI images can be used to diagnose a handful of hemolytic anemias, no effort has been made to infuse it into health care at scale. This project develops an AI system to identify RBCs in crowded QPI images, as well as other blood cellular components, such as platelets and white blood cells. The project will develop robust machine learning models to learn from these data; in particular, new deep learning models, based on forms of convolutional neural networks and recurrent neural networks, will provide insight based on temporal and spatial data.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.

这个小企业技术转让（STTR）第一阶段项目的更广泛影响是通过将3D相位成像与深度学习相结合，提供快速，负担得起和准确的溶血性贫血诊断和监测，从而覆盖服务不足的社区，并解决溶血性贫血患者的医疗保健不平等问题。镰状细胞病（“SCD”）是一种全球性健康问题，其显著影响受影响个体的寿命、生活质量和健康结果。SCD是撒哈拉以南非洲和美国最常见的溶血性疾病之一，影响高达3%的新生儿人口。然而，很少有资源和研究致力于改善SCD的诊断和监测。缺乏筛查和检测的人容易受到高达90%的早期死亡率的影响。不幸的是，那些最有可能患上像SCD这样的溶血性疾病的人往往得不到先进的医疗保健服务。 该项目可通过提供监测以防止需要住院治疗和紧急护理的危机，从而显著减轻SCD的负担。这个小企业技术转让（STTR）第一阶段项目将推进人工智能（AI）创新，通过定量相位成像（QPI）诊断红细胞疾病。虽然QPI图像可用于诊断少数溶血性贫血，但尚未努力将其大规模注入医疗保健。该项目开发了一种人工智能系统，用于识别拥挤的QPI图像中的红细胞，以及其他血细胞成分，如血小板和白色血细胞。该项目将开发强大的机器学习模型，以从这些数据中学习;特别是，基于卷积神经网络和递归神经网络形式的新深度学习模型将提供基于时间和空间数据的洞察力。该奖项反映了NSF的法定使命，并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，被认为值得支持。