Section: New Results

Medical robotics

Needle detection and tracking in 3D ultrasound

Participants : Pierre Chatelain, Alexandre Krupa.

We developed an algorithm for detecting and tracking a flexible needle in a sequence of 3D ultrasound volumes when it is manually inserted, without any a priori information on the insertion direction. Our approach is based on the combination of a RANSAC algorithm with Kalman filtering in a closed loop fashion and allows real-time tracking of the needle. In addition, a pose-based visual servoing was developed for automatically moving a robotized 3D ultrasound probe in order to keep the needle tip centered in the volume and to align its main axis with the central plane of the volume. This needle detection algorithm and probe automatic guidance were experimentally validated during the insertion of a needle in a gelatin phantom [38] .

Non-rigid target tracking in ultrasound images

Participants : Marie Babel, Alexandre Krupa.

In order to robustly track the motion of a tumour or cyst during needle insertion, we developed a new approach to track a deformable target within a sequence of 2D ultrasound images. It is based on a dedicated hierarchical grid interpolation algorithm (HGI) that is typically used for real-time video compression purposes. This approach provides a continuous motion representation of the target by using a grid of control points that models both their global displacement and local deformations. The motion of each control point is estimated by a hierarchical and multi-resolution local search method in order to minimize the sum of squared difference of the target pixel intensity between successive images. This new approach was validated from 2D ultrasound images of real human tissues undergoing rigid and non-rigid deformations.

Adaptive arc-based path planning for robot-assisted needle 3D steering using duty-cycling control technique

Participant : Alexandre Krupa.

This study concerned the development of a method for three dimensional steering of a beveled-tip flexible needle that can be used in medical robotics for percutaneous assistance procedures. The proposed solution is the extension of an adaptive arc-based 2D planar approach. It combines the Rapidly-Exploring Random Tree (RRT) algorithm, the duty-cycling needle control technique and stop and turn phases to reorientate the needle in a new working plane each time it is necessary. Simulation results demonstrate the feasibility of this approach to reach a 3D target while avoiding obstacles and its robustness to needle kinematic model errors.

Gait analysis

Participants : Cyril Joly, Patrick Rives.

Clinical evaluation of frailty in the elderly is the first step to decide the degree of assistance they require. Advances in robotics make it possible to turn a standard assistance device into an augmented device that may enrich the existing tests with new sets of daily measured criteria. We designed an augmented 4-wheeled rollator, equipped with a Kinect and odometers, for daily biomechanical gait analysis. It allows to estimate on line legs and feet configurations during the walk. Preliminary results [43] obtained on four healthy persons show that relevant data can be extracted for gait analysis (e.g. foot orientation and tibia-foot angle, feet position) during an assisted walk.

This work has been realized in collaboration with Claire Dune from the University of Toulon and in the scope of the Inria large-scale initiative action PAL (see Section  8.2.6 ).