Recently, the Prof. Qiushi Ren’s group at Peking University College of Future Technology/Shenzhen Bay Laboratory proposed a new multimodal ocular functional imaging and analysis technology for non-invasive detection of dynamic oxygen metabolism in the ocular nerve and retina. Part of the research results formed the paper "Functional Imaging of Human Retina Using Integrated Multispectral and Laser Speckle Contrast Imaging" published in the Journal of Biophotonics, an international authoritative journal in the field of biomedical photonics.
More than 80% of the information in the human brain is obtained through the eyes, so health of eyes is essential to ensure people's quality of life. At the same time, the human eye is also a natural window for observing brain cognition, human metabolism, human microcirculation, and cardiovascular and cerebrovascular status. Many major chronic diseases may be complicated or secondary to eye diseases. Clinical studies have also confirmed that fundus lesions have the hallmark features of various chronic diseases such as hypertension, stroke, coronary heart disease, neurodegenerative disease, diabetes, and kidney disease, and can be used as a judgment standard for early screening and auxiliary diagnosis of the disease. Therefore, an intelligent multimodal ocular functional imaging and analysis technology, combined with big data and artificial intelligence technology for early warning, screening and monitoring of blinding eye diseases and other chronic diseases can help realize the forward movement of major chronic disease management and is important for saving medical resources and promoting people's health.
The retina has the only directly observable microcirculatory system in the body and one of the most oxygen-consuming organs in the body. Structural and functional information of the retinal is an important guide for disease screening, diagnosis and prognosis. With the development of optical imaging technology, the assessment of retinal has gradually shifted from structural to functional analysis. Retinal multispectral imaging technology not only allows assessment of retinal oxygen saturation, but also enables imaging of retinal structures at different levels for better lesion assessment. On the other hand, non-invasive and large field of view retinal perfusion imaging information can be obtained by laser speckle contrast imaging, which provides effective information for retinal blood supply and hemodynamic analysis. However, existing retinal structural and functional imaging instruments are independent of each other, making it difficult to achieve simultaneous measurement and evaluation, which greatly limits the structural and functional analysis of the retina. Prof. Qiushi Ren's group integrates multispectral imaging technology with laser speckle contrast imaging technology, and combines eye movement analysis and pupil detection technology to propose a novel multimodal ocular functional imaging and analysis technology (shown in Figure 1), which can realize retinal multispectral imaging, fundus color composite image, retinal vessel diameter measurement, retinal oxygen saturation measurement, retinal and choroidal blood perfusion imaging, retinal blood flow pulsation analysis, retinal oxygen metabolism kinetics and other structural and functional information, providing a powerful tool for a more comprehensive assessment of retinal microcirculation characteristics.
Figure 1. The multimodal ocular functional imaging and analysis system (left: imaging modality; right: engineering prototype)
Figure 2 (a) shows the results of 6-wavelength multispectral images (the corresponding wavelengths are marked in white in the image). With increasing wavelengths, the structural information of the retina at different depths can be observed, which is of reference significance for the diagnosis of lesions. In addition, the system provides an expandable light source interface that can be further extended to 12 wavelengths, which is not only helpful for the analysis of structural information in different layers of the retina, but also potentially valuable for the analysis of retinal microcirculatory tissue components. Considering the reading habits of ophthalmologists, this technique combines 470 nm, 550 nm and 600 nm multispectral images into a color composite image, as shown in Fig. 2 (b), and the wavelengths used for the synthesis can be set individually according to the needs of the physicians. Oxygen saturation (the percentage of oxyhemoglobin and deoxyhemoglobin in the blood) is one of the most important indicators to assess oxygenation. At 550 nm, the absorption capacity of oxyhemoglobin and deoxyhemoglobin is approximately equal; at 600 nm, the difference between the absorption capacity of oxyhemoglobin and deoxyhemoglobin is larger. Based on this principle, this technology analysis the retinal oxygen saturation by using 550 nm and 600 nm multispectral images, as shown in Figure 2 (c), in which the closer to red represents a higher oxygen saturation; the closer to blue represents a lower oxygen saturation. Figure 2 (d) shows the mean perfusion images of the subject's retina and choroid, where closer to white represents higher blood flow rate; closer to black represents lower blood flow rate. Figure 2 (e) shows the pulsation curve of retinal blood flow rate in response to heartbeat. The analysis of the pulsation curve not only allows the assessment of heart rate, systolic time, diastolic time etc. but also has potential value for vascular elasticity and hemodynamic analysis. Finally, Figure 2 (f) shows the distribution of retinal blood flow velocity at different moments, which increases and decreases periodically with the heartbeat. The white text in the image identify the different moments, with the closer to red representing the higher blood flow rate and the closer to blue representing the lower blood flow rate.
Figure 2 Structural and functional imaging results. (a) Multispectral photography images; Each image is marked with wavelength; (b) Color fundus composite image; (c) Retinal SO2 mapping calculated from 550 nm and 600 nm images in the MSI sequence. Annular ROI is shown in blue; (d) Laser speckle contrast (LSC) image with long time integration (1s) shows both retinal and choroidal vasculature; (e) Pulsatility waveform generated from blood flow speckle imaging sequence (a 4-cycle part) in the optic disc region, with an obvious signal drop caused by eye movement at the end of the second cycle ; (f) Temporal sequence of laser speckle contrast images; Each image is marked with time stamp.Abbreviations: SO2, blood oxygen saturation; LSC, laser scatter contrast value (1/LSC2 positively correlated with blood flow velocity).
Prof. Qiushi Ren's group has been working on the development of multimodal ocular functional imaging and analysis technology for many years, and the group will continue to research and optimize this technology in the future. The group is currently working with hospitals to conduct large sample clinical studies of multimodal ocular function imaging and analysis technology in ophthalmology, neurology, cardiology, and nephrology, and is expected to conduct more comprehensive research on disease screening, diagnosis, and prognosis, so that the window of the mind can protect health.
This research is supported by the National Biomedical Imaging Facility Grant, National Natural Science Foundation of China, Beijing Natural Science Foundation, Shenzhen Science and Technology Program, and Shenzhen Nanshan Innovation and Business Development Grant.
Article Link.http://doi.org/10.1002/jbio.202100285