Cerebral blood flow (CBF) is the main objective index for clinical diagnosis of cerebrovascular diseases such as cerebral infarction and cerebral hemorrhage. Among them, the measurement of regional cerebral blood flow (rCBF) is of great significance for targeted long-term and real-time detection of target areas of specific diseases such as epilepsy and Alzheimer's disease. In recent years, non-invasive spectral methods for CBF detection have developed rapidly. The more widely employed blood flow monitoring methods are laser speckle contrast imaging (LSCI), diffuse correlation spectroscopy (DCS), interferometric diffusing wave spectroscopy (iDWS), and diffusing speckle contrast analysis (DSCA), which all share the advantage of non-invasive measurement of blood flow adopting non-ionizing radiation. In addition to building an analytical model for detecting rCBF, this paper proposes an interferometric diffusing speckle contrast analysis (iDSCA) method and further constructs an experimental system. The system consists of three modules of laser source module, optical heterodyne module, and imaging acquisition module.

The iDSCA method combines the advantages of iDWS and DSCA, which can achieve high sensitivity and high-resolution two-dimensional velocity imaging, and is of research significance for the long-term detection of rCBF. The electric field intensity of scattered light carries the motion information of scattered particles. In the DSCA principle, the speckle contrast K of diffusing speckle is the integral function of the electric field time autocorrelation function within the exposure time, and it is also the blurring degree of the dynamic speckle image. The reciprocal of its square is employed as the relative blood flow index (BFI) of scattered particles to evaluate the actual blood flow state. Aiming at the measurement error caused by various noise interference in the iDSCA model to calculate the speckle contrast K, the real speckle contrast K obtained by pre-evaluating and correcting the system noise can avoid interference such as laser source noise and camera noise. In this study, the feasibility of this method to detect the linearity of rCBF flow velocity, and the discrimination ability and quantitative analysis ability of this system for different target regions of blood vessels to be measured are verified by analyzing the parameters of multiple diameters and multiple distances through the local phantom flow velocity experiment of the brain. In addition, in vivo experiments and cuff-induced occlusion protocol experiments are carried out at different parts, and blood pressure is measured simultaneously.

The system can effectively improve the signal-to-noise ratio and detection accuracy of non-invasive rCBF detection. The results of phantom flow velocity experiments show that the relative BFI has good linearity with the actual flow velocity, and the average linear correlation coefficient within the source-detector distance (SD) of 6–12 mm is 0.9881 ± 0.0005 (Fig. 6). This detection method can distinguish the flow velocity changes in different target areas, and the relative error of 4.8 mm tube diameter is 2.04% (Table 1). Combined with the vascular diameter measurement method, the flow velocity and flow can be effectively monitored. The increasing trend of BFI measured at SD of 6-12 mm is consistent with the change of flow. The results show that the system can better detect the target area to be measured with a large cross-sectional area within the effective range (Fig. 7). Through in vivo experiments and cuff-induced occlusion protocol experiments (Fig. 8 and Fig. 9), it is proven that the system can detect the flow velocity information of rCBF in vivo and has good detection accuracy within the range of effective measurement flow velocity.

As the optical method for monitoring rCBF is difficult to achieve two-dimensional blood flow imaging, this paper builds a diffusion speckle imaging system with optical heterodyne structure based on the diffusing interference spectrum technology. The improved diffusing speckle contrast analysis method is combined to detect rCBF in real time. Firstly, the feasibility of the system to detect the flow velocity linearity, the discrimination ability, and the quantitative analysis ability of different target areas to be measured are verified through the experimental design of the phantom flow velocity of the brain from analyzing multi-diameter and multi-distance parameters. Secondly, the in vivo experiments of different parts are further designed to verify the measured BFI signals by combining signals of blood pressure. Additionally, the cuff-induced occlusion protocol is conducted to compare the BFI waveforms in three states to verify the reliability of the system detecting and distinguishing rCBF in different regions. This study is expected to achieve non-invasive and long-term monitoring of rCBF and provide a theoretical basis for early diagnosis and treatment of cerebrovascular diseases. In the future, studies will be further conducted on the qualitative and quantitative analysis ability of the system combined with the iDSCA method to detect rCBF, and the clinical application value of rCBF two-dimensional blood flow imaging.