Browsing by Author "Fey, Ann Majewicz"
Now showing 1 - 4 of 4
- Results Per Page
- Sort Options
Item Data-Driven Models for Real-Time Prevention of Adverse Events in Robotic Needle Interventions(2020-09-11) Narayan, Meenakshi; Fey, Ann MajewiczRobotic needle interventions enable surgeons to maneuver flexible needles into deep tissues with enhanced dexterity. The technology shows tremendous potential for therapeutic applications if critical events related to needle-tissue interaction mechanics are addressed during intra-operative procedure. Common events include needle buckling due to flexibility of the shaft, undesired needle deflections due to tissue inhomogeneity, and tissue displacements due to insertion force of the needle. Accurate and timely predictions of these adverse events is important to ensure patient safety. However, it is challenging to provide autonomous control for these events given the unknown dynamic nature of the tissues. The goal of this dissertation is to develop model-free methods to identify, predict and minimize adverse events in real time. The first topic of this dissertation focuses on identifying and detecting the listed three events using only sensor measurements. The algorithms use errors between the sensor readings and estimation models generated from the sensor data to track rapid changes in sensor patterns through scale independent metrics. These metrics enable classification of events based on intensity and behavior. Validation experiments in known gelatin tissues and biological tissues with unknown environments show generalizability. Early prediction of adverse events could improve procedural safety. The second topic of this dissertation focuses on developing computationally fast methods to forecast general time series data, as adverse events are characterized by time series based sensors. This novel forecasting technique is derived from a model-free adaptive control framework. The technique offers high computation efficiency and does not require any parameter tuning or training data, compared to existing statistical and learning models. Forecast accuracy is comparable with current state-of-the art statistical methods. Validations on realistic data sets show application to general time series for data prediction. Using sensor data forecasts and detection metrics from first topic can predict adverse events before occurrence. Preliminary validation experiments in gelatin and biological tissues show robustness to tissue environments. The low computation times of the combined prediction models show potential for integration with autonomous control routines. Dynamics of needle-tissue interactions are unknown and nonlinear. The last topic of the dissertation discusses model-free methods for needle steering control to prevent adverse events during insertions. The traditional model-free adaptive scheme is modified to improve robustness against unexpected changes in system behavior or process errors. Simulation results in known nonlinear systems show consistent stability and faster convergence under unexpected process errors. Experimental results in gelatin tissues show successful prevention of adverse events through prediction and steering control scheme. Results in this dissertation show potential applications to autonomous prediction and control of sensor-driven robotic systems that require safe operations.Item Haptic Stroke Testbed for Pharmacological Evaluation of Dynamic Allodynia in Mouse Models(IEEE Computer Society) Lee, Jin; Atwood, Brian J.; Megat, Salim; Dussor, Gregory; Price, Theodore J.; Fey, Ann Majewicz; Lee, Jin; Atwood, Brian J.; Megat, Salim; Dussor, Gregory; Price, Theodore J.; Fey, Ann MajewiczDynamic mechanical allodynia is an aggravating neuropathological condition in which light, physical touch leads to pain. Developing pharmaceutical agents to treat this condition requires extensive animal trials using a mouse model, and a laborious process of manually stroking inflicted mouse paws, with a brush or cotton swab, while recording responses to that stimulus. In this paper, we developed an autonomous testing mechanism to create repeatable stroking sensations for mice during dynamic allodynia testing. The chamber consists of a belt driven brush mechanism and light and dark chambers. Additionally, we conducted a human subjects study to determine the baseline variability in human-performed dynamic allodynia testing. Our tactile stoke display is capable of stroking a mouse paw between 1-5 mm/s with a repeatable force. In our human subject experiments, the user applied force ranged from 0.1-9.0 gF with a maximum standard deviation of 4.13 gF. In contrast, our device is capable of producing repeatable brush strokes at 0.69 gF (SD = 0.13 gF) and 1.78 gF (SD = 0.16 gF) for two brushes. Preliminary animal studies show that normal mice are not disturbed by the stroking sensation; however, mice afflicted with allodynia move away from it. On average the injured mice spent 90% of their time in a bright, adverse environment to avoid the brush, whereas normal mice only spent 40% of their time in the bright environment.Item Human-Centric Predictive Model of Task Difficulty for Human-In-The-Loop Control Tasks(Public Library of Science) Wang, Ziheng; Fey, Ann Majewicz; Wang, Ziheng; Fey, Ann MajewiczQuantitatively measuring the difficulty of a manipulation task in human-in-the-loop control systems is ill-defined. Currently, systems are typically evaluated through task-specific performance measures and post-experiment user surveys; however, these methods do not capture the real-time experience of human users. In this study, we propose to analyze and predict the difficulty of a bivariate pointing task, with a haptic device interface, using human-centric measurement data in terms of cognition, physical effort, and motion kinematics. Noninvasive sensors were used to record the multimodal response of human user for 14 subjects performing the task. A data-driven approach for predicting task difficulty was implemented based on several task-independent metrics. We compare four possible models for predicting task difficulty to evaluated the roles of the various types of metrics, including: (I) a movement time model, (II) a fusion model using both physiological and kinematic metrics, (III) a model only with kinematic metrics, and (IV) a model only with physiological metrics. The results show significant correlation between task difficulty and the user sensorimotor response. The fusion model, integrating user physiology and motion kinematics, provided the best estimate of task difficulty (R² = 0.927), followed by a model using only kinematic metrics (R² = 0.921). Both models were better predictors of task difficulty than the movement time model (R² = 0.847), derived from Fitt’s law, a well studied difficulty model for human psychomotor control.Item Modeling and Evaluation for Robot-Assisted Surgical Training and Intuitive Teleoperation(2019-12) Wang, Ziheng; 0000-0003-1099-3141 (Wang, Z); Fey, Ann MajewiczWith the advance of teleoperated robotic platforms, such as the da Vinci surgical systems (Intuitive Surgical Inc., Sunnyvale, CA), robot-assisted surgery (RAS) has revolutionized a wide range of surgical interventions towards a safe, reliable, and minimally invasive approach. The objective of this PhD research is to design and develop computational techniques that are capable of automatically measuring a surgeon’s technical skills, interpreting the inherent difficulty demand of complex human motor control tasks, and providing efficient teleoperation evaluation with regard to the human operator. Specifically, the dissertation aims to (1) develop a human-centric analytical approach to objectively evaluate robot-assisted teleoperation, (2) develop machine learning based approaches for automated surgical skill assessment in basic surgical training tasks while improving the interpretability of the assessment to support efficient acquisition of skills, and (3) develop methods for an transferrable and task independent assessment of the operative difficulty demand and surgical skills by extending analysis to broader scenarios of robotic surgical teleoperation. Based on novel techniques of computational intelligence and human-centric analysis, this PhD research highlights the potential to improve surgical training and intuitive design of teleoperation systems for high performing robot-assisted surgery.