Integrated EEG/NIR Sensor System for Infants

描述(申请人提供)：该SBIR项目的目标是设计一种成本效益高、重量轻、集成的全头密集阵列脑电和近红外光谱(DEEG/NIRS)脑成像和数据分析系统，用于非侵入性记录新生儿和幼儿的脑活动。该系统将允许立即对高危新生儿进行床边监测，并及早识别和干预预测发育障碍和认知缺陷的异常。位于头皮表面的由128个组合光电电极组成的密集阵列将同时记录脑电活动(EEG)和脑血氧变化(NIR)，提供大脑功能的补充措施。在第一阶段证明了部分阵列和单一调制波长的可行性后，第二阶段的第一个具体目标是扩展该方法，以完成商业上可行的、全头和双波长的婴儿DEEG/NIRS采集系统。现有的设计将被修改为使用双波长LED发射器来测量血红蛋白，增强了集成光电电极的稳定性和稳健性。新的架构将以不同的频率调制多达的近红外光源(32个双波长LED)，并解调来自多达96个探测器的信号。NIRS数据采集软件将得到完善，并将进一步改进与其他采集流的集成和同步，包括Net Station EEG、Polygraph Input Box Data(例如，ECG、EMG)和E-Prime实验控制软件。第二个具体目标是设计商业、先进的数据分析工具，用于准确计算通过集成的DEEG/NIRS传感器测量的血氧变化。这款近红外光谱分析软件将补充EGI现有的脑电处理和来源分析软件。第一步是实施标准计算工具(基于修正的Beer-Lambert方程)，这些工具用于确定通过恢复的近红外信号测量的氧合和脱氧血红蛋白的变化。然后，通过先进的婴幼儿头部解剖模型，以及近红外光通过婴幼儿头部组织的路径、散射和吸收特性的数值模拟，来精炼计算参数。第三个具体目标将测试和验证集成的DEEG/NIRS系统硬件和软件的数据完整性、功能性和可用性。在30名新生儿的静息状态和功能任务中，将使用完整的128传感器系统同时收集DeEG和NIR数据。功能数据任务将呈现简单的语音和匹配的非语音声音，这些声音之前曾导致脑电模式的个体差异，预测以后的阅读障碍诊断。内部近红外光谱分析工具的输出将与现有的开源分析工具(例如，HOMER)进行交叉验证。将对采购和分析软件的可用性进行评估，并进行进一步改进。最后，将准备硬件、采集软件和分析软件的商业用户指南，以随DEEG/NIRS采集和分析产品一起使用。

DESCRIPTION (provided by applicant): The goal of this SBIR project is to design a cost-effective, lightweight, integrated, whole-head dense-array electroencephalography and near-infrared spectroscopy (dEEG/NIRS) brain imaging and data analysis system for non-invasive recording of brain activity in neonates and young children. This system will permit bedside monitoring of immediate at-risk neonates, and early identification and intervention for abnormalities that predict developmental disabilities and cognitive deficits. A dense array of 128 combined optoelectrodes sitting on the surface of the scalp will simultaneously record brain electrical activity (EEG), and cerebral blood oxygenation changes (NIRS), providing complementary measures of brain function. Having demonstrated feasibility in Phase I with a partial array and single modulated wavelength, the first Specific Aim in Phase II is to extend the approach to complete a commercially viable, whole-head and dual-wavelength dEEG/NIRS acquisition system for infants. The existing design will be modified to use dual-wavelength LED emitters for hemoglobin measurements, with enhanced stability and robustness of the integrated optoelectrodes. New architecture will modulate up to 64 NIR light sources (32 dual-wavelength LEDs) at different frequencies and demodulate signals from up to 96 detectors. NIRS data acquisition software will be refined and further improvements will be made to the integration and synchronization with other acquisition streams, including Net Station EEG, Polygraphic Input Box data (e.g., ECG, EMG), and E-Prime experimental control software. The second Specific Aim is to design commercial, advanced data analysis tools for accurate computation of blood oxygenation changes measured with integrated dEEG/NIRS sensors. This NIRS analysis software will complement EGI's existing EEG processing and source analysis software. The first step will be to implement standard computational tools (based on the modified Beer-Lambert equation) that are used to determine changes in oxy- and deoxy-hemoglobin as measured by recovered NIR signals. Then, the computational parameters will be refined through advanced anatomical infant head modeling, and numerical modeling of the path, scattering, and absorption properties of NIR light through infant head tissues. The third Specific Aim will test and validate the integrated dEEG/NIRS system hardware and software for data integrity, functionality, and usability. Simultaneous dEEG and NIR data will be collected with the complete 128-sensor system during resting state and functional tasks in 30 neonates. The functional data task wil present simple speech and matched non-speech sounds that have previously resulted in individual differences in brain electrical patterns predictive of later dyslxia diagnosis. Outputs from in-house NIRS analysis tools will be cross validated with existing open-source analysis tools (e.g., HomER). Usability of the acquisition and analysis software will be assessed and further refinements made. Finally, commercial user guides for hardware, acquisition software, and analysis software will be prepared to accompany the dEEG/NIRS acquisition and analysis product.