A new human heart vessel identification, segmentation and 3D reconstruction mechanism

Abstract

BackgroundThe identification and segmentation of inhomogeneous image regions is one of the most challenging issues nowadays. The surface vessels of the human heart are important for the surgeons to locate the region where to perform the surgery and to avoid surgical injuries. In addition, such identification, segmentation, and visualisation helps novice surgeons in the training phase of cardiac surgery.MethodsThis article introduces a new mechanism for identifying the position of vessels leading to the performance of surgery by enhancement of the input image. In addition, develop a 3D vessel reconstruction out of a single-view of a real human heart colour image obtained during open-heart surgery.ResultsReduces the time required for locating the vessel region of interest (ROI). The vessel ROI must appear clearly for the surgeons. Furthermore, reduces the time required for training cardiac surgery of the novice surgeons. The 94.42% accuracy rate of the proposed vessel segmentation method using RGB colour space compares to other colour spaces.ConclusionsThe advantage of this mechanism is to help the surgeons to perform surgery in less time, avoid surgical errors, and to reduce surgical effort. Moreover, the proposed technique can reconstruct the 3D vessel model from a single image to facilitate learning of the heart anatomy as well as training of cardiac surgery for the novice surgeons. Furthermore, extensive experiments have been conducted which reveal the superior performance of the proposed mechanism compared to the state of the art methods.Electronic supplementary materialThe online version of this article (doi:10.1186/s13019-014-0161-1) contains supplementary material, which is available to authorized users.

Development and diseases of the coronary microvasculature and its communication with the myocardium

“…Medium sized and smaller coronary vasculature can be visualized in patients, using CT and coronary angiography. Recently significant progress in segmentation methods (automatic identification of coronary vessels) on CT coronary angiography images have been made (Al‐Surmi et al, 2014). Incorporating deep learning methods improved identification of the vessels in vivo, while performing cardiac surgery (Xian et al, 2020; Yang et al, 2019).…”

Tissue clearing and imaging methods for cardiovascular development

“…Rapid improvement is also seen in the field of cardiac surgery image segmentation, where Deep learning is used for segmenting-identifying coronary architecture in CT coronary angiography ( Xian et al, 2020 ; Yang et al., 2019 ). Even before applying Deep learning methods, segmentation methods were used to identify the main vessels and heart chambers when performing an open heart surgery ( Al-Surmi et al., 2014 ; Ecabert et al., 2011 ). Segmentation methods applied onto heart and coronary vasculature and its imaging methods in patients are summarized in Chen et al.…”

Using deep learning to identify the recurrent laryngeal nerve during thyroidectomy

“…However, relatively few works have analyzed open surgery images, where there exists even more complexity and variability in the surgical scene than seen in laparoscopy. For example, in this setting, computer vision has only been studied in heart vessel segmentation in open-heart surgery 41 and wound segmentation in videos of open-neck procedures 42 . The role of computer vision in open surgery remains less-studied than that in endoscopy, laparoscopy, and robotic surgery due to the higher difficulty and often lower quality of image capture during open procedures.…”