Data-Driven Detection of Needle Buckling Events in Robotic Needle Steering

Abstract

In robotic needle steering, flexible asymmetric-tip needles can steer around obstacles to reach targets deep within tissue. Due to tissue inhomogeneity and needle flexibility, needle buckling can occur, preventing accurate placement. This paper focuses on detecting needle buckling using axial force and needle-tip position readings from sensors. Our algorithm uses errors between the force readings and a predictive force model generated from those readings to track rapid changes in the measured forces. Using this prediction error and needle-tip position, the algorithm detects unexpected force increase, strict needle buckling, and buckling with sliding events at the needle-tip. The metrics for the detections are derived using a standard three-sigma rule and a sigmoid function to ensure generalizability of this method to a variety of tissue types. Our algorithm was tested using insertions into a gelatin tissue with an embedded rectangular obstacle designed to elicit buckling events. Needle buckling was detected at a maximum of 2[Formula: see text]mm after collision with the obstacle. Our algorithm was tested for robustness with insertions in an ex vivo tissue under different boundary conditions. Our algorithm was also able to detect buckling events 1–2[Formula: see text]s sooner than human detection times, showing significance for future autonomous control.

Developing a novel force forecasting technique for early prediction of critical events in robotics

“…The first reference force,F � N ðkÞ is a vector of previous N values of the optimal mean forces,f � p generated from the past (k − 1) values of the force data. More details about the procedure to obtainf � p are found in our previous work [21]. Thus,F � N ðkÞ represents the general trend pattern of the previously observed force data.…”

“…The detection steps for axial force increase and needle buckling events are presented with respect to the CFDL-MFP forecasted forces. The complete details of the buckling detection algorithm including the procedure for generating optimal mean force,f � p and the selection of the sigmoid detection metrics, M can be found in [21]. As a first step, the difference betweeñ f a ðkÞ andf � p ðkÞ is normalized using a sigmoid function, g to obtain M:…”

“…In order to lower the risks of this damage, it is crucial to minimize the lag between the time of event occurrence and the time of event detection. We attempted to minimize this lag through our previously developed algorithm that detected needle buckling events within 2mm after encounter with a hard impenetrable obstacle, and at least one second sooner than human detection [21]. However, detecting an adverse event on occurrence can still be too late to prevent any potential damage to the needle-tissue system [8].…”

“…However, detecting an adverse event on occurrence can still be too late to prevent any potential damage to the needle-tissue system [8]. Thus, it is more crucial to predict a buckling event before it occurs so that an appropriate decision can be implemented to prevent the needle from buckling, as illustrated in Fig 1. Since nearly constant needle-tip position and force patterns were used to characterize needle buckling [21], the first step is to perform forecasts of axial force data.…”

“…step forecasts and modifying the existing buckling detection algorithm [21], a prediction algorithm is developed to predict a potential needle buckling at k fi . How soon a prediction is performed before a buckling actually occurs is determined by the prediction time (k b − k fi ).…”