Deep-Learning-Based Cerebral Artery Semantic Segmentation in Neurosurgical Operating Microscope Vision Using Indocyanine Green Fluorescence Videoangiography

There have been few anatomical structure segmentation studies using deep learning. Numbers of training and ground truth images applied were small and the accuracies of which were low or inconsistent. For a surgical video anatomy analysis, various obstacles, including a variable fast-changing view, large deformations, occlusions, low illumination, and inadequate focus occur. In addition, it is difficult and costly to obtain a large and accurate dataset on operational video anatomical structures, including arteries. In this study, we investigated cerebral artery segmentation using an automatic ground-truth generation method. Indocyanine green (ICG) fluorescence intraoperative cerebral videoangiography was used to create a ground-truth dataset mainly for cerebral arteries and partly for cerebral blood vessels, including veins. Four different neural network models were trained using the dataset and compared. Before augmentation, 35,975 training images and 11,266 validation images were used. After augmentation, 260,499 training and 90,129 validation images were used. A Dice score of 79% for cerebral artery segmentation was achieved using the DeepLabv3+ model trained using an automatically generated dataset. Strict validation in different patient groups was conducted. Arteries were also discerned from the veins using the ICG videoangiography phase. We achieved fair accuracy, which demonstrated the appropriateness of the methodology. This study proved the feasibility of operating field view of the cerebral artery segmentation using deep learning, and the effectiveness of the automatic blood vessel ground truth generation method using ICG fluorescence videoangiography. Using this method, computer vision can discern blood vessels and arteries from veins in a neurosurgical microscope field of view. Thus, this technique is essential for neurosurgical field vessel anatomy-based navigation. In addition, surgical assistance, safety, and autonomous surgery neurorobotics that can detect or manipulate cerebral vessels would require computer vision to identify blood vessels and arteries.

Quantitative perfusion assessment of gastric conduit with indocyanine green dye to predict anastomotic leak after esophagectomy

Summary Impaired gastric conduit perfusion is a risk factor for anastomotic leak after esophagectomy. The aim of this study is to evaluate the feasibility of intraoperative quantitative assessment of gastric conduit perfusion with indocyanine green fluorescence angiography as a predictor for cervical esophagogastric anastomotic leak after esophagectomy. Indocyanine green fluorescence angiography using the SPY Elite system was performed in patients undergoing a transhiatal or McKeown esophagectomy from July 2015 through December 2020. Ingress (dye uptake) and Egress (dye exit) at two anatomic landmarks (the tip of a conduit and 5 cm from the tip) were assessed. The collected data in the leak group and no leak group were compared by univariate and multivariable analyses. Of 304 patients who were evaluated, 70 patients developed anastomotic leak (23.0%). There was no significant difference in patients’ demographic between the groups. Ingress Index, which represents a proportion of blood inflow, at both the tip and 5 cm of the conduit was significantly lower in the leak group (17.9 vs. 25.4% [P = 0.011] and 35.9 vs. 44.6% [P = 0.019], respectively). Ingress Time, which represents an estimated time of blood inflow, at 5 cm of the conduit was significantly higher in the leak group (69.9 vs. 57.1 seconds, P = 0.006). Multivariable analysis suggested that these three variables can be used to predict future leak. Variables of gastric conduit perfusion correlated with the incidence of cervical esophagogastric anastomotic leak. Intraoperative measurement of gastric conduit perfusion can be predictive for anastomotic leak following esophagectomy.

Augmented Reality and Image-Guided Robotic Liver Surgery

Artificial intelligence makes surgical resection easier and safer, and, at the same time, can improve oncological results. The robotic system fits perfectly with these more or less diffused technologies, and it seems that this benefit is mutual. In liver surgery, robotic systems help surgeons to localize tumors and improve surgical results with well-defined preoperative planning or increased intraoperative detection. Furthermore, they can balance the absence of tactile feedback and help recognize intrahepatic biliary or vascular structures during parenchymal transection. Some of these systems are well known and are already widely diffused in open and laparoscopic hepatectomies, such as indocyanine green fluorescence or ultrasound-guided resections, whereas other tools, such as Augmented Reality, are far from being standardized because of the high complexity and elevated costs. In this paper, we review all the experiences in the literature on the use of artificial intelligence systems in robotic liver resections, describing all their practical applications and their weaknesses.