LIU Jiaxin， GUI Zhiguo， ZHANG Quan， SHANG Yu
（Shanxi Provincial Key Laboratory for Biomedical Imaging and Big Data, North University of China， Taiyuan 030051, China）
Abstract： By applying the Huber regression algorithm to a relatively new technology of diffuse correlation spectroscopy (DCS), the blood flow index (BFI) from light electric field temporal autocorrelation data is extracted accurately via the Nth-order linear (NL) algorithm. The NL algorithm can extract BFI from tissues with irregular geometric shapes, and its accuracy depends on iterative linear regression. The combination of Huber regression with the NL algorithm is proposed in this paper for the first time. The Huber regression is compared with traditional ordinary least square (OLS) regression through computer simulations for evaluation. The results show that the Huber regression is more accurate in extracting BFI than OLS. Compared to the OLS with an error rate of 4.58%, Huber achieves a much smaller error rate (3.54%), indicating its potential in future clinical applications.
Key words： diffuse correlation spectroscopy (DCS)； blood flow； Huber regression； Nth-order linear (NL) algorithm
References
［1］Shang Y, Li T, Yu G. Clinical applications of near-infrared diffuse correlation spectroscopy and tomography for tissue blood flow monitoring and imaging. Physiological Measurement, 2017, 38(4): 1-26.
［2］Shang Y, Cheng R, Dong L, et al. Cerebral monitoring during carotid endarterectomy using near-infrared diffuse optical spectroscopies and electroencephalogram. Physics in Medicine & Biology, 2011, 56(10): 3015.
［3］Schytz H W, Hansson A, Phillip D, et al. Spontaneous low-frequency oscillations in cerebral vessels: applications in carotid artery disease and ischemic stroke. Journal of Stroke & Cerebrovascular Diseases: the Official Journal of National Stroke Association, 2010, 19(6): 465-474.
［4］Zhou C, Choe R, Shah N, et al. Diffuse optical monitoring of blood flow and oxygenation in human breast cancer during early stages of neoadjuvant chemotherapy. Journal of Biomedical Optics, 2007, 12(5): 1-11.
［5］Choe R, Durduran T. Diffuse optical monitoring of the neoadjuvant breast cancer therapy. IEEE Journal of Selected Topics in Quantum Electronics, 2012, 18(4): 1367-1386.
［6］Hou Y, Shang Y, Cheng R, et al. Obstructive sleep apnea-hypopnea results in significant variations in cerebral hemodynamics detected by diffuse optical spectroscopies. Physiological Measurement, 2014, 35(10): 2135.
［7］Buckley E M, Cook N M, Durduran T, et al. Cerebral hemodynamics in preterm infants during positional intervention measured with diffuse correlation spectroscopy and transcranial Doppler ultrasound. Optics Express, 2009, 17(15): 12571.
［8］Strangman G, Franceschini M A, Boas D A. Factors affecting the accuracy of near-infrared spectroscopy concentration calculations for focal changes in oxygenation parameters. Neuroimage, 2003, 18(4): 865-879.
［9］Boas D A, Campbell L E, Yodh A G. Scattering and imaging with diffusing temporal field correlations. Physical Review Letters, 1995, 75(9): 1855-1858.
［10］Irwin D, Dong L, Shang Y, et al. Influences of tissue absorption and scattering on diffuse correlation spectroscopy blood flow measurements. Biomedical Optics Express, 2011, 2(7): 1969-1985.
［11］Shang Y, Li T, Chen L, et al. Extraction of diffuse correlation spectroscopy flow index by integration of Nth-order linear model with Monte Carlo simulation. Applied Physics Letters, 2014, 104(19): 194.
［12］Shang Y, Yu G. A Nth-order linear algorithm for extracting diffuse correlation spectroscopy blood flow indices in heterogeneous tissues. Applied Physics Letters, 2014, 105(13): 51.
［13］Zhang P. Denoising of N-order linear algorithm for near infrared diffuse light flow measurement. Taiyuan: North University of China, 2019.
［14］Lambert-Lacroix S, Zwald L. Robust regression through the Huber's criterion and adaptive lasso penalty. Electronic Journal of Statistics, 2011, 5: 1015-1053.
［15］Dong L, He L, Lin Y, et al. Simultaneously extracting multiple parameters via fitting one single autocorrelation function curve in diffuse correlation spectroscopy. IEEE Transactions on Biomedical Engineering, 2013, 60(2): 361-368.
［16］Dennis E K. Statistical accuracy in fluorescence correlation spectroscopy. Physical Review A, 1974, 10(6): 1938-1945.
基于Huber回归的漫射相关光谱组织血流测量方法
刘佳鑫， 桂志国， 张 权， 尚 禹
（中北大学 生物医学成像与影像大数据山西省重点实验室， 山西 太原 030051）
摘 要: 将Huber回归算法应用于漫射相关光谱这一相对较新技术中， 通过N阶线性算法(Nth-order linear algorithm, NL algorithm)从光电场时间自相关数据中稳健地提取血流指数（Blood flow index, BFI）。 NL算法可以从不规则几何形状的组织中提取血流指数， 其准确性依赖于迭代线性回归。 本文首次提出将Huber回归与NL算法相结合， 并通过计算机模拟将Huber回归与传统的普通最小二乘法（Ordinary least square, OLS）回归进行了比较。 结果表明， Huber回归提取血流指数比OLS方法更准确。 与误差率为 4.58% 的OLS方法相比， Huber对应的误差率更小(3.54%)， 表明了其未来的临床应用潜力。
关键词: 漫射相关光谱； 血流值； Huber回归； N阶线性算法
引用格式： LIU Jiaxin， GUI Zhiguo， ZHANG Quan， et al. Tissue blood flow measurement by diffuse correlation spectroscopy based on Huber regression. Journal of Measurement Science and Instrumentation, 2021, 12（2）： 127-132. DOI： 10.3969／j.issn.1674-8042.2021.02.001
[full text view]
