Novel Pigment Sensing Pulse Oximeter Technology for Mitigating Racial Bias in Oxygen Saturation Measurements

新型颜料感应脉搏血氧仪技术可减少血氧饱和度测量中的种族偏见

• 批准号: 10599746
• 负责人: Michael McGeehan
• 金额: $ 25.62万
• 依托单位: PENDERIA TECHNOLOGIES, INC.
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-01-01 至 2024-12-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10599746
• 关键词: Adhesives; Admission activity; Adoption; Architecture; Biological; Blood; Blood gas; COVID-19; Caring; Classification; Clinical; Clip; Colorimetry; Compensation; Congresses; Custom; Darkness; Data; Development; Devices; Diagnosis; Disease; Electronics; Ensure; Equitable healthcare; Fingers; Forehead; Functional disorder; Funding; General Anesthesia; Goals; Growth; Health; Health Care Costs; Health Personnel; Healthcare; Hemoglobin; Hospitals; Hypoxemia; Individual; Intensive Care; Interview; Light; Machine Learning; Manufacturer; Marketing; Measurement; Measures; Medical; Medical Device; Medical Technology; Minority Groups; Mission; Modeling; Monitor; Nature; Nurses; Operative Surgical Procedures; Optics; Organ; Oxygen; Oxygen saturation measurement; Patient Care; Patient Monitoring; Patients; Performance; Persons; Phase; Photoplethysmography; Physicians; Physiologic pulse; Pigments; Procedures; Process; Pulse Oximetry; Radial; Research; Research Personnel; Safety; Sampling; Skin; Skin Pigmentation; Sleep Apnea Syndromes; Small Business Technology Transfer Research; Surveys; System; Technical Expertise; Techniques; Technology; Testing; Transducers; Translating; United States; Validation; Ventilator; Work; absorption; career networking; clinical investigation; clinical practice; cohort; commercialization; design; health care disparity; high reward; high risk; human subject; improved; innovation; innovative technologies; light scattering; machine learning algorithm; minority healthcare; mortality; new technology; novel; optical sensor; outcome disparities; people of color; prototype; racial bias; racial disparity; respiratory; sensor; sensor technology; signal processing; skin color; standard measure; success; technology development; wearable device; wireless electronic

Abstract Wearable pulse oximetry sensors are among the most ubiquitous medical technologies used in healthcare. Clinicians rely on pulse oximeters to monitor patient health during disease states and medical procedures. However, numerous studies have shown this technology to be racially biased; hypoxemia is three times more likely to go undetected in patients with dark skin compared to those with light skin. This technological deficit disproportionately impacts people of color in the United States and contributes to minority healthcare disparities. The FDA has since issued a safety warning about using pulse oximetry for minority populations. Low blood oxygen levels can signify serious respiratory illness in a patient and healthcare workers rely on pulse oximeters to determine what type of care a patient needs. False negative hypoxemia diagnosis can have detrimental consequences for patient care including increased rates of mortality and organ dysfunction. The widespread and impactful use of these devices highlights a critical need for innovative technology that improves healthcare equity and mitigates racial bias in pulse oximetry. Despite well-documented racial biases, a 2022 survey of pulse oximeter technology found no manufacturers attempting to incorporate novel technology that accounts for differences in skin tone. The goal of this Phase I project is to develop novel sensor technology for non-invasively measuring blood oxygen saturation that accounts for the patient’s skin pigment and delivers reliably accurate oxygen saturation measurements for persons of all skin colors. Our proposed technology incorporates two sensing modules: a novel skin pigment colorimetry sensor for classifying patient skin tone and an optical transducer for measuring blood oxygen saturation via the photoplethysmography technique. We will develop a machine learning algorithm to model and compensate for the effects of skin tone on blood oxygen saturation measurements. These innovations in technology development and biological signal processing will meet a critical need for a medical device that provides accurate blood oxygen saturation monitoring for persons of all skin colors.

抽象的 可穿戴脉冲血氧饱和传感器是医疗保健中使用的最普遍的医疗技术之一。 临床医生依靠脉搏氧合在疾病状态和医疗程序中监测患者健康。 但是，许多研究表明这项技术大致有偏见。低氧血症多三倍 与皮肤浅色的患者相比，皮肤深色皮肤的患者可能不会发现。这种技术防御 不成比例地影响美国有色人种，并导致少数族裔医疗保健差异。 此后，FDA发布了有关将脉搏氧用于少数族裔人群使用的安全警告。低血氧 水平可以表示患者的严重呼吸道疾病，医护人员依靠脉搏氧合来 确定患者需要哪种类型的护理。假阴性低氧诊断可能有害 患者护理的后果，包括增加死亡率和器官功能障碍。宽度和 这些设备的有影响力的使用强调了对创新技术的关键需求，以改善医疗保健权益 并减轻脉搏血氧饱和度中的种族偏见。尽管有据可查的种族偏见，但2022年的脉搏调查 血氧仪技术发现没有制造商试图合并新的技术来说明 肤色差异。该阶段I项目的目标是开发非侵入性的新型传感器技术 测量血氧饱和度，以说明患者的皮肤色素并可靠地提供 所有肤色的人的氧气安全测量。我们提出的技术结合了两个 传感模块：一种新型的皮肤色素色素法，用于分类患者肤色和光学 通过光摄像学技术测量血氧安全性的传感器。我们将发展一个 机器学习算法以建模并补偿肤色对血氧饱和的影响 测量。这些技术开发和生物信号处理方面的创新将达到关键 需要为所有肤色的人提供准确的血氧饱和度监测的医疗设备。