Tendon Strain Apparatus

Our client has a mechanical bioreactor that is used to test the mechanical properties of engineered tendon constructs and tendon tissues in his lab. His system is able to measure tensile force applied and displacement between the grips holding the tissue. We researched and bought a camera, lens, and camera mount and created a non-contacting video system through LabVIEW to track the strain of soft biological tissues in our clients lab.

Problem Definition
Tendons are collagenous tissues that transfer forces from muscle to bone. Tendons have poor healing capabilities and can be easily injured due to overuse. In order to develop therapies for preventing injury and improving regeneration more knowledge must be gained on how mechanical forces influence tendon cell behavior. Our client has designed and built a mechanical bioreactor controlled by Labview to test mechanical properties of soft tissues such as tendon. Currently the client is able to measure the force applied to the tissue and the displacement between the grips holding the tissue. Our objective is to develop and design a non-contacting video system to measure the strain within soft biologic tissues in real time and improve the capability of the system already in place.

Project Requirements and Goals
 * Camera resolution: less than 1 mm (between 0.5 mm and 1 cm is optimal)
 * LabView must link to and utilize the non-contacting video system
 * The system will be placed inside an incubator so all components must be hardened to humidity, moisture, and heat
 * Our system must work for the length of time that a test will run which can be multiple hours

Background
Tendons are the tissues are responsible for transferring forces between muscle and bone. Tendon injuries are quite common among athletes, with causes ranging from chronic overuse to acute tears. The regenerative process for tendons can take over a year and often complete regeneration does not occur. This has led to a demand in clinical alternatives for replacement. ​Before mechanical tissue replacements can be engineered, more data is needed on the mechanical and biological factors that govern tendon injury and regeneration.

Project Learning
As our project progresses in development, we solve one problem after another. The following is a categorical breakdown of the project learning for each major conceptual milestone.

Camera
Originally, we wanted to do a GoPro type of camera. After the first Snapshot and talking with our client we decided a CMOS or CCD camera would be more logical. Through researching, we found that CMOS cameras would most likely be the way to go.

Below is the comparison of cameras we have researched or found used in similar video systems. We have listed what we believe to be some of the more important factors of the camera system. As you can see, they are only a few minor details that differ per camera.

Camera research is an aspect we continued to work on. As a team, we decided the camera system is one of our top priorities and this will most likely be our biggest expense. We were hopeful, and succeeded, to have a camera selected and purchased before the beginning of the second semester. This will allowed us to move onto the next design phase of implementing the video system with the existing apparatus and computer programs.

Programing
Matlab

During the camera-to-Matlab integration process, the team spent some time getting to know the image capture and image processing toolboxes that Matlab provides (for a price, of course). With these toolboxes, the camera is able to digitally subtract pixel from pixel and mathematically determine the difference in pixel change.

Additionally, the team spent more time becoming comfortable with Matlab terminology and procedural equations.

Labview



Fall Semester Progress: The Labview code was given to us by our client and currently controls the load cell that can stretch the tendon at a constant load or with cyclic loading.

The interface for the MATLAB integration module is shown to the right. This module will allow a user to utilize MATLAB code without leaving the LABview window. One goal of this project is to create a seamless, straightforward system to allow for efficient data acquisition. Once complete, the LABview code combined with the integrated MATLAB code will be able partly control the apparatus, stream a live feed of the tissue being tested, and calculate the strain being applied over a time interval.

Spring Semester Progress: Our programming had come along far enough that we are able to track the movement of a marked point by our design review on February 2. We were also able to plot the movement onto an x,y plot. Before purchasing our camera we were using a webcam. With the webcam, we were able to identify two different marked points but was not able to plot the two points separately. We were working on tracking both points independently of one another on a strain plot. Another goal at the time was to allow for reference distance points and to have the LabView code find those reference points as well.

Mounting
Some specifications for mounting the camera to the frame of the bioreactor include:
 * The mounting system needs to be able to move the camera out of the way when the user is preparing the sample
 * The camera needs to come back to a fixed location for image acquisition
 * The mounting device needs to be able to support the weight of the camera

Our team will begin to put more emphasis on a mounting design in the second semester of senior design. We will test different potential materials for encasing the camera in the incubator to see if there is condensation formed that could interfere with the image.

Final Product
Camera and Lens

After careful consideration and evaluation we have purchased the THORLABS Compact DCC1645C CMOS camera(shown below left) that is listed in the first column of the camera comparison table above.We chose this camera because it is compact, lightweight, has a small pixel size, and a high resolution. It is also easy to use and connects to the computer using a USB.

We were able to purchase a lens that we were able to previously test out thanks to our lead instructor Dev Shrestha. This lens (shown below right) features an 8mm focal length allowing us to not have any distortion errors within our live feed as well as a manual focus to make sure it is as focused as the researchers choose. With these two components together, we were able to accomplish a compact, cost efficient, and effective camera system that can be incorporated into our programming system.

Matlab

We spent time looking over both softwares and the capabilities allotted in each. After talking about it with our instructor and as a team we decided to focus our code in LabVIEW and not use MATLAB. We made this decision because it helped streamline our code and it has a relatively user friendly graphical user interface.

LabView

Our non-contact video system displays the live video feed in the left window (shown below) while simultaneously measuring the displacement of two points on a collagen sponge or tendon. This displacement is then calculated and graphed in the window to the right of the video feed. We placed indicators the show the numerical value of strain as the system is running, the displacement, and the number of point matches the program is picking up. We also have two error indicators which help the user to easily find what and where the errors are if there happens to be any. Shown below in Figure 6 is the block diagram of our LabVIEW program. This is the behind the scenes of our programing. The live video feed comes into the ‘vision acquisition’ module. From there the image is pulled into a module called ‘vision assistant’. Within this module there are many different options for image analysis. To track the change in length of the tissue we used ‘pattern matching’. Once the two points on the soft tissue are found, our code outputs the x-values of each point. This is used to find the displacement. The smaller loop above our main loop also tracks the displacement between the two dots but it stops after it gets one data point. This gives us the initial displacement between the two points. We use the initial displacement and the changing displacement to get our strain calculation over time. To convert this calculation from pixels to millimeters we use another ‘vision assistant’ module with a feature called 'guage' to create a millimeter to pixel ratio that's unique to each time the system runs, depending on camera placement. Using this ratio, we convert our strain calculation to millimeters and it is plotted into the chart displayed on the front panel.

Mounting

The mount we have chosen for our project (shown right) is an articulating arm with center lock produced by Tether Tools. The mount is 7” long and is capable holding devices up to 4.5 lbs. The arm has ¼”-20 male threads on both ends of it, enabling it to connect to any ¼”-20 female receptor. The arm comes with a removable hot shoe adapter which can be used to attach it to a camera. One requirement for the mount was that it must be easy to adjust to any position inside the incubator. The articulating arm by Tether Tools can be moved quickly and easily and can be locked by turning a knob at the center of the arm.

Resources and Documentation
Meeting Minutes [[Media:Meeting Minutes Dev-elopers.pdf|Fall Semester]] [[Media:2016-17_Dev-elopers_Meeting Minutes Spring.pdf|Spring Semester]]

EXPO [[Media:EXPO Poster Dev-elopers.pdf|EXPO Poster]] [[Media:2016-17_Dev-elopers_EXPO Presentation.pdf|EXPO Technical Presentation]]

Other Documents [[Media:2016-17_Dev-elopers_Final Report.pdf|Final Report]] [[Media:Team Contract Dev-elopers.pdf|Team Contract]]