Nine years ago, Julian Gunderson was born three months prematurely and weighed just 1 kg (2.2 lb). After a year's worth of injections, medications, transfusions, and other treatments, doctors diagnosed Julian with cerebral palsy.
Cerebral palsy is a non-progressive but incurable neurological disorder that affects millions of children around the world. It is the result of brain lesions that occur while the child is still in the mother's womb, or shortly after birth, often impacting the child's muscular control. The term non-progressive indicates that the condition does not get worse over time, and physical therapy treatments early in life can help children with cerebral palsy improve their movement and function. Beyond that, there are very few medical training devices available today that are directly used for treating cerebral palsy.
However, for inventor Marc Gunderson — Julian's father — the term non-progressive is something he refuses to take at face value. Drawing his drive and inspiration from his son, Marc has developed an innovative line of electronic medical training devices and games geared toward improving his son's balance and posture.
Branded as the JPosture collection, many of Marc's devices incorporate FlexiForce™ sensors — an ultra-thin force-sensitive resistor technology (FSR) — as a critical component to capture force-driven biofeedback data and benchmark his performance over time. Ranging from more simple and intuitive training tools, to more complex, Bluetooth-enabled systems, the JPosture devices are useful and engaging methods to aggressively treat cerebral palsy.
Application 1: J-Shoes
The J-shoes were developed to track the weight distribution between a subject's feet to evaluate and improve balance. The goal of the J-Shoes is to remind the subject to stand correctly if they forget.
Each shoe includes four FSRs embedded into the sole — two positioned at the front of the foot, and two under the heel. In the current prototype, each shoe is fitted with an Arduino Micros microcontroller to capture the force feedback from the FSRs, but the shoe on the right foot is the main controller.
The main controller receives raw force readings from the left foot's sensors and compares them with the right foot values. If the two sets of values remain within an acceptable difference in balance from one another, nothing happens. However, if either set's force readings go above an acceptable difference after a programmed time period, the main controller will trigger a vibrating mechanism on the left shoe, indicating that the user should correct their balance.
The shoes communicate with each other via Bluetooth. The left shoe has one Bluetooth module (HC-05) that sends the force application data to the main controller on the right shoe. The right shoe, however, has two Bluetooth modules — one communicating with the left foot's microcontroller, and the other communicating with an Android phone application.
The phone app not only receives balance data from the shoes, it sets the parameters for the real-time force monitoring system. The app user can set the acceptable percentage of balance difference between the two feet and also the amount of time the subject can stand out-of-balance until the vibrating mechanism is triggered.
While the J-Shoes are currently programmed to address Julian's specific standing/weight-bearing challenges, Marc plans to create different codes to address a wider range of balance and stability conditions.
Application 2: Finger Trainer
About a year ago, Julian decided he wanted to play the piano. It became apparent that his cerebral palsy affected his finger motor coordination, making it difficult for him to play certain chords. As a result, the Finger Trainer was developed to exercise Julian's finger motor coordination. The Figure Trainer is comprised of five individual strips incorporating single-point FSRs linked to a series of eight programmable LED lights protected inside a frosted acrylic tube. Each FSR is covered with colored foam that correlates to the color of the LED light with-in each strip. In this current prototype, an Arduino Uno converts the applied force captured from the FSR and displays the feedback in the series of LED lights. As more force is applied, the more lights will light up in the respective tube.
Going forward, Marc envisions programming challenges into the device for the subject to make certain finger movements at different applied forces, along with a method to score or grade their performance. He also plans to advance the design from individual tubes into a force-sensitive wearable glove, which would not require subjects to reposition the force-sensitive strips each time they lift their hands.
Application 3: Two Weight-Bearing Platform Games
Given the fact that the JPosture devices are meant for children, Marc has developed two different games that can only be played when the subject is exhibiting proper body posture and balance.
In both games, a wireless, pressure-sensitive platform is used to track the subject's weight distribution and stance while the game is being played. The platform is made of two movable foot panels. Each foot panel has four large FSRs; just like the J-Shoes, there are two FSRs are positioned at the front of the foot, and two at the heel.
A knob on the platform can be calibrated for a “sway-allowance variation” between 10 and 50 percent of the total difference of pressure between the two feet. If the platform is set to a 10 percent sway-allowance setting (the strictest possible setting), this means the game will not allow subjects to proceed until they balance themselves within a 10 percent pressure threshold between their two feet. Likewise, with the setting at 50 percent, subjects can be leaning twice as heavily on either side of their stance, and the game will still proceed.
In his first game prototype, Marc instrumented a toy gun with a laser to wirelessly communicate with the weight-bearing platform through two Series-2 XBee modules. The laser light on the gun will only trigger when the subject is balanced within the sway-allowance percentage setting. When proper posture and weight distribution is established, the subject can shoot at a target that has 15 individual LEDs linked with 15 photo resistors (controlled by an Arduino Mega2560).
As the game is played, one of the 15 LEDs will randomly light up for three seconds at a time. The subject is instructed to shoot the laser gun at the lit LED. If the subject targets the lit LED with the laser correctly — while maintaining balance — a photoresistor linked to the LED senses the laser gun's light, and a four-digit display score counter will begin increasing. If the subject deviates off of the sway-allowance setting at any point during the game, the laser will not output until the subject corrects his or her balance. This challenges the subject's ability to keep balanced while testing hand-eye coordination.
In a separate game prototype, the same pressure-sensitive weight-bearing pad is linked to a marble run course. The goal of the marble run game is to get the marbles successfully down a series of eight tracks and into the proper collection bin, while avoiding the two escape shoots that are positioned on either side of the structure.
Every few seconds, when the subject is standing in proper balance, a dispensing servo wirelessly activates and opens a block that covers a hole at the top track of the run. At the end of each track, the marble hits a wall and drops onto the next track. However, at the end of tracks three through six, there are four hatches leading to the escape shoots that remain closed only when proper balance is detected.
If subjects do not maintain their balance within the sway-allowance setting, the servo-driven hatches will open, and the marble will fall down the escape shoot. Points are awarded via a four-digit score counter for proper standing, while points are removed for off-balance stances.
With Progress Made, the Drive Continues for Father and Son
These are just a few of the smart inventions Marc has developed to challenge cerebral palsy conditions head-on. He hopes to collaborate with like-minded inventors seeking the same goals he and Julian are targeting. He is constantly improving and refining his devices and continues to explore new opportunities with creative and effective medical training innovations.
This article was written by Andy Dambeck, Content Specialist for Tekscan, Inc., South Boston, MA. To connect with Marc Gunderson about his inventions, visit here. For more information on Tekscan, visit here.