OPTICAL NAVIGATION TECHNIQUES FOR MINIMALLY INVASIVE ROBOTIC SURGERIES

Optical imaging system;Endoscopic Imaging;Surgical endoscopy;3D reconstruction;Robotic Navigation;Minimally Invasive Robotic Surgery;Structured Illumination;Fringe Projection Profilometry;Multispectral Imaging;Point cloud Analysis;Surgical Planning;Brain Imaging;In vivo experiments;Photoacoustic Imaging;Fluorescence Microscopy;Multimodal Imaging.;Electrical Engineering

Le, Hanh NDKang, Jin Ung ;

Johns Hopkins University

关键词: Optical imaging system;    Endoscopic Imaging;    Surgical endoscopy;    3D reconstruction;    Robotic Navigation;    Minimally Invasive Robotic Surgery;    Structured Illumination;    Fringe Projection Profilometry;    Multispectral Imaging;    Point cloud Analysis;    Surgical Planning;    Brain Imaging;    In vivo experiments;    Photoacoustic Imaging;    Fluorescence Microscopy;    Multimodal Imaging.;    Electrical Engineering;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/60118/Figure3.18B_Video.mp4?sequence=4&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

Minimally invasive surgery (MIS) involves small incisions in a patient;;s body, leading to reduced medical risk and shorter hospital stays compared to open surgeries.For these reasons, MIS has experienced increased demand across different types of surgery. MIS sometimes utilizes robotic instruments to complement human surgical manipulation to achieve higher precision than can be obtained with traditional surgeries. Modern surgical robots perform within a master-slave paradigm, in which a robotic slave replicates the control gestures emanating from a master tool manipulated by a human surgeon. Presently, certain human errors due to hand tremors or unintended acts are moderately compensated at the tool manipulation console. However, errors due to robotic vision and display to the surgeon are not equivalently addressed.Current vision capabilities within the master-slave robotic paradigm are supported by perceptual vision through a limited binocular view, which considerably impacts the hand-eye coordination of the surgeon and provides no quantitative geometric localization for robot targeting. These limitations lead to unexpected surgical outcomes, and longer operating times compared to open surgery.To improve vision capabilities within an endoscopic setting, we designed and built several image guided robotic systems, which obtained sub-millimeter accuracy. With this improved accuracy, we developed a corresponding surgical planning method for robotic automation. As a demonstration, we prototyped an autonomous electro-surgical robot that employed quantitative 3D structural reconstruction with near infrared registering and tissue classification methods to localize optimal targeting and suturing points for minimally invasive surgery.Results from validation of the cooperative control and registration between the vision system in a series of in vivo and in vitro experiments are presented and the potential enhancement to autonomous robotic minimally invasive surgery by utilizing our technique will be discussed.