Instruments For Robotic Surgery Research Articles

Accurate instance segmentation of surgical instruments in robotic surgery: model refinement and cross-dataset evaluation.

Automatic segmentation of surgical instruments in robot-assisted minimally invasive surgery plays a fundamental role in improving context awareness. In this work, we present an instance segmentation model based on refined Mask R-CNN for accurately segmenting the instruments as well as identifying their types. We re-formulate the instrument segmentation task as an instance segmentation task. Then we optimize the Mask R-CNN with anchor optimization and improved Region Proposal Network for instrument segmentation. Moreover, we perform cross-dataset evaluation with different sampling strategies. We evaluate our model on a public dataset of the MICCAI 2017 Endoscopic Vision Challenge with two segmentation tasks, and both achieve new state-of-the-art performance. Besides, cross-dataset training improved the performance on both segmentation tasks compared with those tested on the public dataset. Results demonstrate the effectiveness of the proposed instance segmentation network for surgical instruments segmentation. Cross-dataset evaluation shows our instance segmentation model presents certain cross-dataset generalization capability, and cross-dataset training can significantly improve the segmentation performance. Our empirical study also provides guidance on how to allocate the annotation cost for surgeons while labelling a new dataset in practice.

An Assessment of Comparative Marginal Costs to Non-Robotic Surgery for Radical Prostatectomy amongst Public Patients

Objective: Surgical procedures assisted by robotic technology have been increasing in uptake, particularly in private hospitals. Unlike most of the studies on robotic technology which were sponsored by pecuniary interests of the dominant Monopoly supplier, this paper is an independent investigators’ initiated study of comparative health system costs in a Local Health District on Robotic Surgery (RS) versus Non-RS for patients undergoing prostatectomy amongst a cohort of public patients. The report was limited to understanding the cost implications of RS and productivity savings associated with it via consistently reduced length of stay (LOS). Patient benefits were not formally measured or assessed. Methods: Estimated marginal costs of both RS and Non-RS were ascertained via cost modelling Activity Based Funding (ABF) cost bucket categories for each inpatient separation for Non-RS radical prostatectomy and comparing these to actual costs for patients undergoing RS for radical prostatectomy. This approach compared the expected non-theatre costs of both RS and Non-RS, and considered appropriate cost bucket comparators for operating theatre costs. Results: The results are in congruence with established literature, RS costs are higher than Non-RS. Specifically, this study found that RS yielded a productivity saving to the health system of two days shorter LOS, than Non-RS. However, the marginal cost per separation of RS prostatectomy is $3086 higher than Non-RS prostatectomy in the instruments and consumables costs. Any potential savings associated with reduced LOS do not offset the higher marginal costs associated with RS. Conclusion: Instruments and consumables for RS remain as monopoly products, save for competitive forces; the health system costs for RS are likely to remain significantly higher than non-RS.

ST-MTL: Spatio-Temporal multitask learning model to predict scanpath while tracking instruments in robotic surgery.

Representation learning of the task-oriented attention while tracking instrument holds vast potential in image-guided robotic surgery. Incorporating cognitive ability to automate the camera control enables the surgeon to concentrate more on dealing with surgical instruments. The objective is to reduce the operation time and facilitate the surgery for both surgeons and patients. We propose an end-to-end trainable Spatio-Temporal Multi-Task Learning (ST-MTL) model with a shared encoder and spatio-temporal decoders for the real-time surgical instrument segmentation and task-oriented saliency detection. In the MTL model of shared-parameters, optimizing multiple loss functions into a convergence point is still an open challenge. We tackle the problem with a novel asynchronous spatio-temporal optimization (ASTO) technique by calculating independent gradients for each decoder. We also design a competitive squeeze and excitation unit by casting a skip connection that retains weak features, excites strong features, and performs dynamic spatial and channel-wise feature recalibration. To capture better long term spatio-temporal dependencies, we enhance the long-short term memory (LSTM) module by concatenating high-level encoder features of consecutive frames. We also introduce Sinkhorn regularized loss to enhance task-oriented saliency detection by preserving computational efficiency. We generate the task-aware saliency maps and scanpath of the instruments on the dataset of the MICCAI 2017 robotic instrument segmentation challenge. Compared to the state-of-the-art segmentation and saliency methods, our model outperforms most of the evaluation metrics and produces an outstanding performance in the challenge.

Multiscale matters for part segmentation of instruments in robotic surgery

Multiscale matters for part segmentation of instruments in robotic surgery

Performance Evaluation of Optically Sensorized Tendons for Articulate Surgical Instruments

Abstract This study proposes an optically sensorized force-sensing tendon for minimally invasive surgical instruments. The tendon is composed of a high strength, polarization maintaining (PM) optical fiber with Bragg sensors (FBGs) that negate the cross-sensitivity of conventional FBGs. The PM-FBG fiber is locally reinforced with high stiffness Kevlar that enhances its load carrying capacity while enabling higher curvatures in tendon routing. The composite tendon has a mean diameter of ∼268 μm which preserves the form-factor of instruments within this scope. Importantly, the tendons can improve the functionality of such tools by enabling local force and tissue-resistance estimation. This paper explores the performance of these sensorized tendons in terms of strength, stability, response under dynamic load, friction, and sensitivity as a force measuring tool within an 18 Ga articulate Nitinol (NiTi) cannula (a proxy for potential applications). Results reaffirm the potential of a bi-modal sensing and actuation component within instruments for robotic surgery.