CAREER: Multiplexing Biomedical Sensor Systems for Automated Diagnosis by Machine Learning

由于人体生理的复杂性，人类健康状况的诊断一直具有挑战性。因此，疾病的诊断和预后往往需要多种检测来识别多种生物标志物。这一过程需要时间和资源，包括训练有素的卫生服务提供者、进行检测的实验室技术人员和解释结果的临床医生。为了解决这一问题，该项目将研究能够同时超灵敏检测多种生物标志物的多路生物医学传感器，以便在早期阶段识别健康问题。该项目还将研究自动化数据分析和诊断系统，利用机器学习算法对测试结果进行准确、公正的解释。最后，该项目将研究与教育和推广计划相结合，旨在激发K-12学生对STEM机会的兴趣，并培养生物传感器、机器学习和纳米技术领域的本科生和研究生，用于生物医学管理。这个CAREER项目是通过协同集成新的生物标志物检测传感平台，以及机器学习算法，开发用于自动诊断的综合生物医学传感器，以准确和公正地解释结果。将追求四个主要目标：(1)研究和制造多激子结构，其中多激子是等离子体-激子-等离子体混合系统，具有适合多路传感的新能量状态；(2)通过探索具有多种生物标志物的多激子结构开发多路传感系统；(3)开发将多种生物标志物的传感器结果关联起来的算法；(4)深入了解多分子结构中的光-物质相互作用以及医学应用中的机器学习。这种方法比传统的方法，如等离子体传感器和ELISA，不提供灵敏度和多路复用同时具有优势。该技术还将通过自动诊断推进即时护理系统。本职业计划的教育部分旨在为纳米工程结构开发光物质相互作用的教育计划。这项工作将通过在实验室中使用纳米结构来激发学生的学习，展示生物医学传感器的多学科视角，并通过生物传感器项目激发学生的协作学习。这项活动将加强STEM领域代表性不足群体的招聘和保留，并将激励学生终身学习和从事与生物医学学科、计算机科学和工程相关的职业。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The diagnosis of human health conditions has always been challenging due to the complexity of human physiology. Therefore, multiple tests to identify multiple biomarkers are often required for diagnosis and prognosis of diseases. This process takes time and resources, including well-trained health providers, laboratory technicians to run the tests, and clinicians to interpret the results. To address this concern, this project will investigate multiplexed biomedical sensors that are capable of ultrasensitive detection of multiple biomarkers simultaneously so that health concerns could be identified in their early stages. The project will also investigate automated data analytics and diagnosis systems by utilizing machine learning algorithms for an accurate and unbiased interpretation of the test results. Finally, this project integrates research with educational and outreach initiatives designed to excite K-12 students about STEM opportunities and train undergraduate and graduate students in the area of biosensors, machine learning, and nanotechnology for biomedical stewardship.This CAREER project is to develop integrative biomedical sensors for automated diagnosis by synergistic integration of new sensing platforms for biomarker detection, as well as machine learning algorithms for accurate and unbiased interpretation of the results. Four main aims will be pursued: (1) study and fabricate plexcimonic structures, where plexcimons are plasmon-exciton-plasmon hybrid systems exhibiting new energy states that are suitable for multiplexed sensing, (2) develop a multiplexed sensing system by exploring plexcimonic structures with multiple biomarkers, (3) develop algorithms that correlate the sensor results for multiple biomarkers, and (4) create an advanced understanding of light-matter interactions in plexcimonic structures and machine learning in medical applications. This approach offers advantages over the conventional methods such as plasmonic sensors and ELISA that do not provide sensitivity and multiplexing simultaneously. The technology will also advance Point-of-Care systems through automated diagnosis. The educational component of this CAREER plan aims to develop an education program for light-matter interactions for nano-engineered structures. This effort will energize student learning by using nanostructures in laboratories, demonstrate a multidisciplinary perspective of biomedical sensors, and stimulate students' collaborative learning through biosensor projects. This activity will enhance recruitment and retention of underrepresented groups in STEM fields and will motivate students towards lifelong learning and careers related to biomedical disciplines, computer science, and engineering in general.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.