ABSTRACT
BACKGROUND AND OBJECTIVE: In order to be context-aware, computer-assisted surgical systems require accurate, real-time automatic surgical workflow recognition. In the past several years, surgical video has been the most commonly-used modality for surgical workflow recognition. But with the democratization of robot-assisted surgery, new modalities, such as kinematics, are now accessible. Some previous methods use these new modalities as input for their models, but their added value has rarely been studied. This paper presents the design and results of the "PEg TRAnsfer Workflow recognition" (PETRAW) challenge with the objective of developing surgical workflow recognition methods based on one or more modalities and studying their added value. METHODS: The PETRAW challenge included a data set of 150 peg transfer sequences performed on a virtual simulator. This data set included videos, kinematic data, semantic segmentation data, and annotations, which described the workflow at three levels of granularity: phase, step, and activity. Five tasks were proposed to the participants: three were related to the recognition at all granularities simultaneously using a single modality, and two addressed the recognition using multiple modalities. The mean application-dependent balanced accuracy (AD-Accuracy) was used as an evaluation metric to take into account class balance and is more clinically relevant than a frame-by-frame score. RESULTS: Seven teams participated in at least one task with four participating in every task. The best results were obtained by combining video and kinematic data (AD-Accuracy of between 93% and 90% for the four teams that participated in all tasks). CONCLUSION: The improvement of surgical workflow recognition methods using multiple modalities compared with unimodal methods was significant for all teams. However, the longer execution time required for video/kinematic-based methods(compared to only kinematic-based methods) must be considered. Indeed, one must ask if it is wise to increase computing time by 2000 to 20,000% only to increase accuracy by 3%. The PETRAW data set is publicly available at www.synapse.org/PETRAW to encourage further research in surgical workflow recognition.
Subject(s)
Algorithms , Robotic Surgical Procedures , Humans , Workflow , Robotic Surgical Procedures/methodsABSTRACT
Objective:To develop a novel, flexible, dual-arm, master-slave digestive endoscopic minimally invasive surgical robot system named dual-arm robotic endoscopic assistant for minimally invasive surgery (DREAMS) and to evaluate its feasibility for endoscopic submucosal dissection (ESD) by using ex vivo porcine stomachs.Methods:A novel endoscopic robot (DREAMS) system was developed which was composed of a flexible two-channel endoscope, two flexible robotic manipulators, a master controller, a robotic arm, and a control system. A total of 10 artificial round-like lesions with diameters ranging from 15 to 25 mm were created (5 in gastric antrum and 5 in gastric body) by using fresh peeled stomach of healthy pigs as the model. Submucosal dissection was performed with the assistance of the DREAMS system by two operators. The main outcome was submucosal dissection speed, and the secondary outcomes included muscular injury rate, perforation rate, and grasping efficiency of the robot.Results:All 10 lesions were successfully dissected en bloc by using the DREAMS system. The diameter of the artificial lesions was 22.34±2.39 mm, dissection time was 15.00±8.90 min, submucosal dissection speed was 141.79±79.12 mm 2/min, and the number of tractions required by each ESD was 4.2 times. Muscular injury occurred in 4/10 cases of ESD. No perforation occurred. Conclusion:The initial animal experiment shows the DREAMS system is safe and effective.
ABSTRACT
Objective:Previous studies have shown an association between programmed death-ligand 1 expression(PD-L1)in non-small cell lung cancer(NSCLC)and clinical factors and that PD-L1 is positively correlated with TNM staging.This study aimed to explore the prognostic significance of PD-L1 and its correlation with the maximum standardized uptake value(SUVmax).Methods:Clinicopath-ological data and the follow-up information of the 122 de novo primary NSCLC patients were analyzed.PD-L1 expression was detected by immunohistochemistry in this 122 surgically resected non-small cell lung carcinoma tissues.Survival outcomes were analyzed using the Kaplan-Meier method and multivariate Cox proportional hazards model.Correlation between SUVmax and PD-L1 expression was analyzed using Spearman's rank correlation analysis.Results:Multivariate analysis revealed that PD-L1 expression(HR=4.518,95% CI:1.176-17.352,P=0.028)and tumor size(HR=1.404,95%CI:1.020-1.933,P=0.037)were independent risk factors for overall survival(OS) in early NSCLC patients.Sex,age,pathological type,CEA level,and SUVmax group had no obvious effect on OS(P 0.05)in early NSCLC patients.In univariate analyses,sex,pathological type,tumor size,and SUVmax group affected OS in stageⅢ-ⅣNSCLC patients.How-ever,age,CEA level,and PD-L1 expression had no effect on OS.PD-L1 expression was not an independent risk factor for OS in stageⅢ-ⅣNSCLC patients.The SUVmax group had no association with PD-L1 in all patients.Conclusions:PD-L1 expression is an independent risk factor for OS in early NSCLC patients but not in stageⅢ-Ⅳpatients.
ABSTRACT
In this paper, a novel finite time fault tolerant control (FTC) is proposed for uncertain robot manipulators with actuator faults. First, a finite time passive FTC (PFTC) based on a robust nonsingular fast terminal sliding mode control (NFTSMC) is investigated. Be analyzed for addressing the disadvantages of the PFTC, an AFTC are then investigated by combining NFTSMC with a simple fault diagnosis scheme. In this scheme, an online fault estimation algorithm based on time delay estimation (TDE) is proposed to approximate actuator faults. The estimated fault information is used to detect, isolate, and accommodate the effect of the faults in the system. Then, a robust AFTC law is established by combining the obtained fault information and a robust NFTSMC. Finally, a high-order sliding mode (HOSM) control based on super-twisting algorithm is employed to eliminate the chattering. In comparison to the PFTC and other state-of-the-art approaches, the proposed AFTC scheme possess several advantages such as high precision, strong robustness, no singularity, less chattering, and fast finite-time convergence due to the combined NFTSMC and HOSM control, and requires no prior knowledge of the fault due to TDE-based fault estimation. Finally, simulation results are obtained to verify the effectiveness of the proposed strategy.
ABSTRACT
Design of low-cost, miniature, lightweight, ultra low-power, flexible sensor platform capable of customization and seamless integration into a wireless biomedical sensor network(WBSN) for health monitoring applications presents one of the most challenging tasks. In this paper, we propose a WBSN node platform featuring an ultra low-power microcontroller, an IEEE 802.15.4 compatible transceiver, and a flexible expansion connector. The proposed solution promises a cost-effective, flexible platform that allows easy customization, energy-efficient computation and communication. The development of a common platform for multiple physical sensors will increase reuse and alleviate costs of transition to a new generation of sensors. As a case study, we present an implementation of an ECG (Electrocardiogram) sensor.