Bai Jing, Zhu Qisen, Liu Yinqiu, Zhou Yihang, Shi Tianlei, Gui Zhiguo, Shang Yu
Shanxi Provincial Key Laboratory for Biomedical Imaging and Big Data, North University of China, No.3 Xueyuan Road, Taiyuan 030051, China.
Shanxi Provincial Key Laboratory for Biomedical Imaging and Big Data, North University of China, No.3 Xueyuan Road, Taiyuan 030051, China.
Comput Methods Programs Biomed. 2020 Sep;193:105456. doi: 10.1016/j.cmpb.2020.105456. Epub 2020 Mar 17.
Tissue blood oxygenation contains critical information for biomedical studies and healthcare. The primary approach to extract the absolute value of tissue blood oxygenation (e.g., oxygen saturation) is spatial-resolved algorithm for near-infrared diffuse optical spectroscopy with continues-wave (CW) light, which require acquisition of the optical signals from multiple pairs of sources and detectors (S-D). This study reports the first attempt for absolute oxygenation measurement with single S-D pair of optical signals.
A novel algorithm, namely, phantom-validation modified Beer-Lambert law (PV-MBLL), was created to fully utilize the optical signals from single S-D pair. This algorithm is combined with two-step phantom measurement to extract the absolute value of tissue oxygenation in CW system. The proposed PV-MBLL algorithm was compared with the conventional spatial-resolved algorithm on both step-varied liquid phantom and human experiment of cuff occlusion on arms. The one-way ANOVA analysis was performed to investigate the difference between the two algorithms.
By using the PV-MBLL algorithm, the reconstructed tissue absorption coefficient is highly accurate (not larger than 5.35% in error) over a wide range (0.02-0.20 cm). By contrast, the spatial-resolved algorithm leads to much larger errors (up to 37.57% in error). Moreover, the responses of oxygen saturation to cuff occlusion differ significantly (p < 0.005) with the two algorithms.
The proposed PV-MBLL algorithm has promising potential for accurate acquisition of oxygenation information. Additionally, the single S-D pair greatly reduces the size of optical probe and instrument cost, thus it is highly appropriate for the tissues with small size and large curvature.
组织血液氧合作用包含生物医学研究和医疗保健的关键信息。获取组织血液氧合绝对值（如氧饱和度）的主要方法是用于连续波（CW）光近红外漫射光谱的空间分辨算法，该算法需要从多对光源和探测器（S-D）采集光信号。本研究首次尝试使用单对S-D光信号进行绝对氧合测量。
创建了一种新算法，即体模验证修正比尔-朗伯定律（PV-MBLL），以充分利用单对S-D光信号。该算法与两步体模测量相结合，以提取连续波系统中组织氧合的绝对值。将所提出的PV-MBLL算法与传统空间分辨算法在阶跃变化液体体模和手臂袖带阻断人体实验上进行比较。进行单因素方差分析以研究两种算法之间的差异。
使用PV-MBLL算法，重建的组织吸收系数在较宽范围（0.02 - 0.20厘米）内具有高度准确性（误差不大于5.35%）。相比之下，空间分辨算法导致的误差要大得多（误差高达37.57%）。此外，两种算法下氧饱和度对袖带阻断的反应差异显著（p < 0.005）。
所提出的PV-MBLL算法在准确获取氧合信息方面具有广阔的潜力。此外，单对S-D极大地减小了光学探头的尺寸和仪器成本，因此非常适合于小尺寸和大曲率的组织。
