Design and validation of looping assistance methods in robotic-assisted neonatal surgical suturing in a chest model.

BACKGROUND: Neonate patients have a reduced thoracic cavity, making thoracoscopic procedures even more challenging than their adult counterparts. METHODS: We evaluated five control strategies for robot-assisted thoracoscopic surgical looping in simulations and experiments with a physical robotic system in a neonate surgical phantom. The strategies are composed of state-of-the-art constrained optimization and a novel looping force feedback term. RESULTS: All control strategies allowed users to successfully perform looping. A user study in simulation showed that the proposed strategy was superior in terms of Physical demand p < 0.05 $\left(p< 0.05\right)$ and task duration p < 0.05 $\left(p< 0.05\right)$ . The cumulative sum analysis of inexperienced users shows that the proposed looping force feedback can speed up the learning. Results with surgeons did not show a significant difference among control strategies. CONCLUSIONS: Assistive strategies in looping show promise and further work is needed to extend these benefits to other subtasks in robot-aided surgical suturing.

The effects of different levels of realism on the training of CNNs with only synthetic images for the semantic segmentation of robotic instruments in a head phantom.

PURPOSE: The manual generation of training data for the semantic segmentation of medical images using deep neural networks is a time-consuming and error-prone task. In this paper, we investigate the effect of different levels of realism on the training of deep neural networks for semantic segmentation of robotic instruments. An interactive virtual-reality environment was developed to generate synthetic images for robot-aided endoscopic surgery. In contrast with earlier works, we use physically based rendering for increased realism. METHODS: Using a virtual reality simulator that replicates our robotic setup, three synthetic image databases with an increasing level of realism were generated: flat, basic, and realistic (using the physically-based rendering). Each of those databases was used to train 20 instances of a UNet-based semantic-segmentation deep-learning model. The networks trained with only synthetic images were evaluated on the segmentation of 160 endoscopic images of a phantom. The networks were compared using the Dwass-Steel-Critchlow-Fligner nonparametric test. RESULTS: Our results show that the levels of realism increased the mean intersection-over-union (mIoU) of the networks on endoscopic images of a phantom ([Formula: see text]). The median mIoU values were 0.235 for the flat dataset, 0.458 for the basic, and 0.729 for the realistic. All the networks trained with synthetic images outperformed naive classifiers. Moreover, in an ablation study, we show that the mIoU of physically based rendering is superior to texture mapping ([Formula: see text]) of the instrument (0.606), the background (0.685), and the background and instruments combined (0.672). CONCLUSIONS: Using physical-based rendering to generate synthetic images is an effective approach to improve the training of neural networks for the semantic segmentation of surgical instruments in endoscopic images. Our results show that this strategy can be an essential step in the broad applicability of deep neural networks in semantic segmentation tasks and help bridge the domain gap in machine learning.

Virtual reality simulation of robotic transsphenoidal brain tumor resection: Evaluating dynamic motion scaling in a master-slave system.

BACKGROUND: Integrating simulators with robotic surgical procedures could assist in designing and testing of novel robotic control algorithms and further enhance patient-specific pre-operative planning and training for robotic surgeries. METHODS: A virtual reality simulator, developed to perform the transsphenoidal resection of pituitary gland tumours, tested the usability of robotic interfaces and control algorithms. It used position-based dynamics to allow soft-tissue deformation and resection with haptic feedback; dynamic motion scaling control was also incorporated into the simulator. RESULTS: Neurosurgeons and residents performed the surgery under constant and dynamic motion scaling conditions (CMS vs DMS). DMS increased dexterity and reduced the risk of damage to healthy brain tissue. Post-experimental questionnaires indicated that the system was well-evaluated by experts. CONCLUSION: The simulator was intuitively and realistically operated. It increased the safety and accuracy of the procedure without affecting intervention time. Future research can investigate incorporating this simulation into a real micro-surgical robotic system.