Intraoperative Tissue Biomechanics Sensor

image from Medical Plastics News

Link to Executive Summary: https://docs.google.com/document/d/1ee8TzNvXldFV7r2OlQj7u8z5U_XfVV83b7TQnBnUa6g/edit?usp=sharing

Sponsored by: Dr. Sameer Shah and the Department of Orthopaedic Surgery and Bioengineering School of Medicine

Background: Despite being the crucial highway of communication in the body, the nervous system hosts a delicate set of tissues that are susceptible to damage. Should these nerves be damaged, repair of the nerves would require analysis of the target area. Currently, surgeons are trained to analyze the damage "by feel", sensing the tension inside these nerves through touching and prodding as a way to gauge the type of procedure that is required. As advanced as medical procedures have become, it is a surprise that the sense of touch is relied on like it is. Dr. Sameer Shah has hypothesized that finding a way to get quantitative values for this tension in the form of stress and strain might prove to be a great aid in the success of the procedures of nerve repair as well as the repeatability of these procedures.

Objective: With that goal set, it has become the task of this project to develop a device that can be used by surgeons to receive real time values of stress and strain in nerve tissue. The hope is that this device (or set of devices) can be used during operations, and ultimately provide data that can help with reproducibility and success of these procedures.

Sponsor objectives included:

• Primary: CAD model and circuit of potential device(s)

• Secondary: Physical model along with some testing on nerve substitute

• "Wow!": Ex vivo testing on animal tissue

Ultimately, the team was able to achieve two separate devices for stress and strain, as well as successfully creating two physical models that were tested on a variety or rubber bands and o-rings that stood to substitute nerve tissue.

Final Stress Design:

The final stress device is designed to attach to the nerve at the end piece, thus allowing for translational movement to come into contact with a force sensitive resistor, ultimately displaying a force reading for the force that exists in the nerve.

While it may not be producing stress values, this concept has been discussed with and approved by the sponsor. It was important to be able to produce accurate force values before implementing the stress model. Said stress model was included in the final report, detailing exactly how stress can be extracted from the force within a nerve.

Left: Breakdown of device components I Right: Gif showing translational movement of device

In the end, the device was able to provide the necessary force readings, calibrated to every 0.5 N, but able to supply readings more specific than that decimal point. After the final iteration of testing, a force-strain curve was produced. The most important success is the fact that the translational concept works, and would only need to be refined in terms of sensitivity.

More detailed results and explanations can be found in the "Final Design" page.

Final Strain Design: A non-contact optical method of analysis called Digital Image Correlation (DIC) was chosen in order to obtain the strain values. Optical methods are generally more attractive because they are completely non-invasive, which makes them 100% biocompatible. They additionally minimize the need to have physical contact with the tissues and potentially damage them.

The strain device uses a camera to record the movement of markers on the tissue’s surface before and after elongation. The image files tracking the markers’ movement are then transferred to a Matlab script that is able to calculate the strain values based on the relative X, Y and Z displacements of the markers. This Matlab software was custom built and is adapted from many other DIC softwares that were tested out. The customized software inputs the recording from a single camera and identifies markers of various radii placed on the tissue. It then calculates the change in apparent radius of the markers in addition to their 2 dimensional X and Y displacement to provide real strain values across the plane. Doing so, the most difficult task of calculating strain values, i.e. when the tissue is stretched at different angles, is addressed.

Experimental testing setup for strain measurement procedure.

The Matlab script inputs videos captured by the camera and breaks it into multiple frames. It then processes these frames individually to measure the location and size of the markers in terms of pixels.