Results

Introduction

The trend of surgical robots is moving away from the traditional master-slave robots and going toward miniaturized endoluminal devices especially for screening and interventions in the gastrointestinal (GI) tract. For example, capsule endoscopy has been performed worldwide in the last five years with successful outcomes. In order to enhance the dexterity of the commercial endoscopic tools, many endoscopic devices have been developed.
These devices are promising for the future development of minimally invasive and endoluminal surgery. However, the implementable functions are generally limited in a single endoscopic capsule because of space constraints. In addition, advance endoscopes with some miniaturized arms have poor dexterity; the diameter of such arms must be very small to be inserted through an endoscopic channel (2-4 mm in general), resulting in a small force generated at the tip.
In the framework of ARES project, a reconfigurable modular robotic system is proposed to overcome the intrinsic limitations of a single-capsule or endoscopic approach and to perform screening and interventions in the GI tract. In the proposed system, miniaturized robotic modules are ingested and assembled in the stomach cavity. The assembled robot can change its configuration according to the target location and target task. The modular approach facilitates the delivery of more components inside a body cavity having small entrance and exit. The robotic reconfiguration allows the precise positioning of the vision and therapeutic tools which can cooperate to perform complicated surgical tasks.

Proposed procedures

The following figure shows the surgical procedures for the proposed modular robot. Prior to the surgical procedure, the patient drinks a liquid to distend the stomach up to 1400 ml. Next, the patient ingests 10-15 robotic modules (SSSA, ETH, UB), and then the swallowed modules complete the assembling process before the liquid naturally drains away from the stomach, which is in 10-20 minutes. Magnetic self-assembly using permanent magnets has been chosen since its feasibility has been demonstrated in this ARES project (ETH). Soon after the assembly, the robot configures its topology as planned based on a preoperative planning (INRIA) by repeated docking and undocking of the modules. The assembly, the robotic configuration and the surgical tasks are controlled via wireless bidirectional communication with an external console operated by the surgeon. Additional modules having different interventional functions and/or modules containing an extra battery can be added later to the robotic structure even during the operation. The modules after use can be detached and discarded if it’s not necessary in the following procedures. In the same way, the module can be easily replaced with the new one when it is broken. After the surgical tasks are completed, the robot reconfigures itself to a snake-like shape, for example, to pass through the pyloric sphincter or completely disassembles itself into individual modules. One of the modules can bring a biopsy tissue sample out of the body for the detailed examinations afterwards (UB).

Robotic module design (SSSA)

The following figure shows the concept of modular topology and design of the structural module. The module has 2 DOFs (±90° of bending and 360° of rotation) and contains a Li-Po? battery, two brushless DC motors and a custom-made motor control board capable of wireless control. Two permanent are attached at each end of the module to help with self-alignment and modular docking. The hexagonal shape at each end restricts the rotational motion after being docked. The adequate choice of the magnets is being investigated to maximize the possibility of self-assembly.
The module size is 15.4 mm in diameter and 36.5 mm in length, that is still to be miniaturized, and the total weight is 5.6g. The figure also shows the design and the prototype of the biopsy module used as a functional module. The grasping parts can be hidden in the module casing to miniaturize its size and to avoid tissue damage during ingestion. Some mock-ups modules have been fabricated to show that the robot can configure into different topologies.

Magnetic Self-Assembly and Connection(ETH)

Due to their long-range passive interaction, the use of magnets as the driving mechanism of selfassembly　seems self-evident: if magnets are placed on the mating faces, the magnetic force will　attract the different modules toward each other and the magnetic torque will orient them. We　introduced the concept of the MAgnetic Self-aligning Hermaphroditic (MASH) mating face and argued that, for a unique connection, the mating faces should present exactly two opposite magnetic poles.
Self-assembly of snake-type robots has been also demonstrated by preattaching magnetic cylindrical joints to the modules. These joints can allow the robot to adapt itself to the GI tract as it moves through it while still maintaining intermodule connection. Using a specific magnet configuration on the connection face, assembly success rates of up to 90% are possible.

Topology planning (INRIA)

Techniques are developed in this project to evaluate (1) the achievable workspace of a kinematic chain where a set of constraints are satisfied, (2) the topology of the kinematic chains where a set of constraints are satisfied for all points in the desired workspace. Numerical algorithm are developed based on the interval analysis techniques. The technique evaluates all continuous range of values in the workspace and the design space, instead of point to point basis, and it is developed to address the workspace evaluation and design generation of general case manipulators.

In addressing the reconfigurable (modular) robots, it was found that tremendous computational advantage can be derived through the simplification of the problems as provided by the physical constraints of the modular robots: link lengths and offset lengths of resulting kinematic chains can only be in multiples of module length, module symmetry and the finite ways of docking orientation reduces the possible combinations of kinematic chains, etc. This allows the technique to reduce the interval variables that need to be evaluated as continuous range of real values and replace them with real variables or interval variables of very narrow widths (to reflect uncertainties in the variables). The resulting algorithm was observed to solve for the topology of a 6DOF serial chain with reconfigurable modules in less than 6 seconds. Computational time would also depends on other factors, such as resolutions of the desired results and the number (and complexity) of the constraints involved.

Biosensor (UB)

The integration of diagnostic tools able to perform in-situ tests giving real time or close to real-time information while performing the surgery would be one of the more visionary goals of this project. In particular, chemical sensors and biosensors could be a great step in in-situ diagnostics, thus overcoming the limitations of the just morphological analysis that is currently provided by optical methods.
In particular, a sensor for pH monitoring would be of special interest for the real-time detection of metabolic tissue activity (see figure below, left image). Regarding surgical microtools, inputs from the medical side have leaded to explore the possibility of adding a tiny micropipette tool able to obtain and store microbiopsy samples (see figure below, right image).



Biocompatibility safety issues are a major concern when designing such a complex robot as the one envision in the ARES project. All the components of the robot have been carefully selected for being already used in medical components and small parts have being checked for citotoxicity to ensure their safe use in the surgical procedures.


ARES:Assembling Reconfigurable Endoluminal Surgical system
CFP6-2003-NEST-B-3, No.015653, 01/02/2006-31/07/2009