Providing accurate image-guidance for soft-tissue interventions remains a complex task. Most of the time, preoperative models and planning data are no more valid during the surgical process due to motions and deformations of the organ of interest. In this paper, two core components of a computer-assisted system for liver surgery are presented. One is an ultrasound segmentation techniques that allows for automatic liver vessels detection and the other is a mass-spring based deformable model used to update the shape of 3D models. Both have real-time capabilities and enable to update intraoperatively the data created during the planning phase.

This paper presents a new approach for estimating physical properties of deformable models from experimental measurements. In contrast to most previous work, we introduce a new method based on particle filters which identifies the different stiffness properties for spring-based models. This approach addresses some important limitations encountered with gradient descent techniques which often converge towards ill solutions or remain fixed in local minima conditions.

We propose an architecture for a real-time multimodal system, which provides non-contact, adaptive user interfacing for Computer-Assisted Surgery (CAS). The system, called M/ORIS (for Medical/Operating Room Interaction System) combines gesture interpretation as an explicit interaction modality with continuous, real-time monitoring of the surgical activity in order to automatically address the surgeon's needs. Such a system will help reduce a surgeon's workload and operation time. This paper focuses on the proposed activity monitoring aspect of M/ORIS. We analyze the issues of Human-Computer Interaction in an OR based on real-world case studies. We then describe how we intend to address these issues by combining a surgical procedure description with parameters gathered from vision-based surgeon tracking and other OR sensors (e.g. tool trackers). We called this approach Scenario-based Activity Monitoring (SAM). We finally present preliminary results, including a non-contact mouse interface for surgical navigation systems.

Mass-spring systems have widely and effectively been used for modeling in real-time deformable objects. Easier to implement and faster than finite elements, these systems, on the other side, suffer from several drawbacks when coming to render physically believable behaviors. Neither isotropic or anisotropic materials can be controlled easily and the large number of springs and mass points composing the model makes it fastidious to define parameters to control elongation, flexion and torsion at a macroscopic level. Another weakness is that most of the materials found in nature maintain a constant or quasi-constant volume during deformations; unfortunately, mass-spring models do not have this property. In this paper, we extend the current state-of-the-art in soft tissue simulation by introducing a six-degree of freedom macroscopic elastic sphere described by mass, inertia and volumetric properties. Spheres are placed along the medial axis transform of the object whose centers are connected by a skeleton composed of a set of three-dimensional elastic links. Spheres represent internal mass, volume and control the global deformation of the object. The surface is modeled by setting point masses on the mesh nodes and damped springs on the mesh edges. These nodes are connected to the skeleton by individual elastic links, which control volume conservation and transfer forces between the surface and volumetric model. Using this framework we also present an efficient method to approximate collision detection between multiple bodies in real-time.

This paper describes the development of efficient computer vision techniques for human-computer interaction. Our approach combines range and color information to achieve efficient, robust tracking using consumer-level computer hardware and cameras. In this paper, we present the design of the Human Oriented Tracking (HOT) library, present our initial results, and describe our current efforts to improve HOT's performance through model-based tracking.

Our goal is to make vehicle teleoperation accessible to all users. To do this, we develop easy-to-use yet capable Web tools which enable efficient, robust teleoperation in unknown and unstructured environments. Web-based teleoperation, however, raises many research issues, as well as prohibiting the use of traditional approaches. Thus, it is essential to develop new methods which minimize bandwidth usage, which provide sensor fusion displays, and which optimize human-computer interaction. We believe that existing systems do not adequately address these issues and have severely limited capability and performance as a result. In this paper we present a system design for safe and reliable Web-based vehicle teleoperation, describe an active and dynamic user interface, and explain how our approach differs from existing systems.

Precise control of the structure of matter at the nanometer scale will have revolutionary implications for science and technology. Nanoelectromechanical systems (NEMS) will be extremely small and fast, and have applications that range from cell repair to ultrastrong materials. This paper describes the first steps towards the construction of NEMS by assembling nanometer-scale objects using a scanning probe microscope as a robot. Our research takes an interdisciplinary approach that combines knowledge of macrorobotics and computer science with the chemistry and physics of phenomena at the nanoscale. We present experimental results that show how to construct arbitrary patterns of gold nanoparticles on a mica or silicon substrate, and describe the underlying technology. We also discuss the next steps in our research, which are aimed at producing connected structures in the plane, and eventually three-dimensional nanostructures.

The potato operation is a project belonging to the domain of agricultural robotics, or agrotics. It aims to automate pulp sampling of potatoes in order to detect viral diseases. The difficulty of this problem lies in the high variability (shape, texture, colour) of natural objects. Three approaches have been studied, based on contour classification, on thermometry and on active vision and image analysis. Cooperation between robotics, active 3D computer vision and robust image analysis leads to a performant solution.>