可视化
生物医学工程
图像分辨率
材料科学
计算机视觉
计算机科学
光学
人工智能
医学
物理
作者
Li Wang,Di Ke,Haishu Xin,Rui Liu,Shu Pan,Kedi Xiong,Sihua Yang
摘要
The current preoperative vascular imaging methods cannot achieve noninvasive high-resolution imaging of deep-localized vessels. Photoacoustic tomography (PAT) can show microvessels with centimeter depth and submillimeter diameter without the use of contrast agents. Combined with PAT and optical projection technology, the Hessian-matrix-based skin removal algorithm and the target matching method were developed to spatially align the photoacoustic data of subcutaneous blood vessels with the anatomy of real patients and to realize three-dimensional (3D) visualization of blood vessels from the body surface. The optical projection navigation system based on PAT has high spatial resolution (∼135 μm) and temporal resolution (0.1 s). In the rabbit injection experiment, 3D distributions of needle and blood vessel (>100 μm) were obtained by image segmentation, which proved that the method can guide micro plastic injection. Furthermore, healthy volunteers' forehead imaging experiments show that 3D visualization and cross-sectional images of the human forehead clearly show the vascular network and ability of the system to image submillimeter blood vessels with penetration depth (∼10.2 mm). Our work confirms that the method of integrated photoacoustic imaging and optical projection has great potential for noninvasive diagnosis and treatment of clinical blood vessels, opening a path for the application of photonics in medical esthetics.
科研通智能强力驱动
Strongly Powered by AbleSci AI