Shaurya Agarwal, Allison Garza, Sonia Garcia, Vivaswath Kumar, Andrew Schober
Cerebral Palsy (CP) is a congenital non-progressive disorder that causes physical disabilities in motor development through damage to the motor control centers of the brain, and affects over 10,000 infants each year. The disability most commonly affects the upper extremities, and requires physical therapy or corrective surgical procedures following extensive evaluation of the patient to try and improve the condition. However, evaluations are typically subjective and difficult to carry out uniformly across patients due to the lack of standardization of upper extremity motion. We are an interdisciplinary team of engineering students from Rice University who have designed an innovative evaluation tool that can quantitatively assess the motion path of individuals that have upper extremity impairments in dexterity and motor capabilities. With this device, physicians and physical therapists will have a means to quantitatively and accurately assess patients’ limitations before surgery and their progress and improvement over time after surgery and therapy.
A standardized quantitative evaluation tool is required by physicians to inform them on the effectiveness of CP patient surgery and therapy regiments. However, even though several evaluation tools such as Gait Analysis methods exist for lower extremities, no standardized quantitative method of evaluation for the upper extremities currently exists. The Functional Dexterity Test (FDT) and Shriners Hospital’s Upper Extremity Evaluation (SHUEE) tool provide qualitative methods to test patients, but fail to provide physicians with data that tracks the improvement of the patient’s quality of motion over time. The scope of the design challenge is to create an evaluation tool for upper extremities that is comparable in quality to current lower extremity quantitative testing methods, yet is a portable and inexpensive option.
Our main objective is to design and prototype a device that more accurately and effectively analyzes the improvement of patients with CP after surgery or therapy over time. The primary design criteria for the device are that it be an accurate, precise, portable and inexpensive tool for comprehensive and quantitative analysis of manual dexterity. As a result, the completed device will produce valuable information regarding the progress and rehabilitation of CP patients without laboratory or video testing. The solution will quantitatively track and analyze the motion path of children with CP in order to improve the quality of rehabilitation, and provide a numerical comparison of this data against standards to track patient improvement over time.
The target population for this product is children ages 6 to 18 with spastic CP. The product technology involves using a small, hollow, plastic cylindrical peg 46mm in length and 28mm in diameter that is easy for children to grip. The peg serves as a housing mechanism for an accelerometer, gyroscope, and magnetometer housed in an Integrated Measurement Unit that is responsible for sensing and capturing motion. To transfer the collected motion path data, a Bluetooth Low Energy chip wirelessly transmits the data to the physician’s computer where the information is then filtered for noise and drift, analyzed, and displayed visually via graphs and charts on an intuitive graphical user interface (GUI) for the physician. Additional components such as a smart battery chip, a current limiter and voltage regulators are also included to improve and ensure the functionality and safety of the device. A standard lithium ion battery powers this chip, and will be charged when the peg, which contains an externally accessible micro-USB port, is connected to any micro-USB charger.
The team additionally designed two exams to test for motion smoothness and accuracy, the two most important characteristics of dexterity for CP patient evaluation according to Dr. Gloria Gogola, a pedatric orthopedic surgeon from Shriners Hospital in Houston, TX with whom the team is closely collaborating. To test patients’ smoothness, the team designed an exam with a series of predetermined paths the patient is instructed to trace with the peg. The patient’s movement is then scored against an average of ideal motions generated from a fully functional, unaffected hand to provide a quantitative evaluation and to track improvement over time. Using the second exam to test accuracy, patients are asked to rotate the peg 180 degrees in their hand while moving the peg over a 4’’ hurdle.
The peg starts in the center of one target, and patients aim to land the peg in the center of the adjacent target after moving it over the hurdle. Both the exams are on two sides of a single rectangular apparatus that can fold onto itself for ease of storage and portability.
Lastly, the final piece of our solution is the Extended Kalman filter. Accelerometers accrue drift as total time of the test increases during each trial.
The extended Kalman filter addresses this problem by using the accelerometer, gyroscope, and magnetometer data to factor out the drift and residual noise to produce a motion path. The extended Kalman Filter is the golden standard for filtering motion data, and will provide the data accuracy and precision needed to track the metrics required by physicians. Filtered data will be analyzed for a variety of metrics, most importantly smoothness and accuracy, and displayed on the computer with comparisons to control patients and also the patient’s progress over time. Altogether, this solution will give the physician a holistic view of the patient’s status, the patient’s improvement over time, and what next steps should be taken for continued improvement.
Each segment of the final solution has been prototyped, tested, and iterated. The prototype for the peg was developed using an Arduino UNO to connect the 9-axis integrated measurement unit (IMU), smart battery chip, voltage regulator and bluetooth module. The extensive amount of testing and alternatives analysis to choose the correct components led to further compatibility testing between the chips. The chips successfully functioned together with the Arduino prototype, and streamed 9-axis IMU data to the computer. The next prototype iteration is to test these chips on the Printed Circuit Board (PCB) before final integration into the peg.
The exam apparatus had a first iteration prototype created by painting the targets and paths on two sides of a 12” x 18” neoprene rubber mat with 3D printed hurdles. The exams were tested with Shriners Hospital patients, who found the exams enjoyable and easy to learn while they provided the team with the data necessary to score accuracy and smoothness with the patient’s motion. The second iteration of the prototype is an opaque acrylic mat with the paths and targets indented into the mat using the etching feature of a precise laser cutter. The mat will have 3D printed clips with Delrin rods for the hurdles, and hinges between the hurdles to allow the mat to fold into a third of the original size for storage and portability.
Moreover, the extended Kalman Filter has yielded positive results by significantly reducing the noise and drift from the IMU data. The team is currently taking additional measures with the filter to continue progressing towards this goal, including adding calibration procedures for the exam, calibrating the hardware itself, filtering out noise from neighboring devices, and adding a low pass filter.
The current solution is designed for use in a hospital or rehabilitation clinic, and is tailored for children by using a simple, colorful apparatus that is easy to set up and intuitive to use. The shape and dimensions of the peg are ideal for a pediatric patient, and the wireless data transfer simplifies physician interaction. The future step for the device is its establishment as a standard pediatric evaluation tool that can provide accurate, quantitative data for all types of motor impairments or disorders, including sports injuries and other upper extremity ailments. This information can be then used to determine whether a patient is in need of surgery, or how they’re progressing through their recovery phase after surgery. In the event that surgery is not required for the patient, the captured data can then support the extent to which physical therapy improves their range of motion. As a result, the primary client for the device is pediatric orthopedic surgeons and physical therapists throughout the United States. Assuming that 80% of these pediatric orthopedic physicians in the United States will be interested in purchasing the product, the team estimates $250,000 in revenue within the first year the product is on the market, and $25,000 per year after that. Individually, the product will cost $5,000 to purchase by the end-consumer, which is all-inclusive in providing the physical product and in lifetime services and maintenance.
Through the use of the device, physicians will be able to accurately measure improvement among CP patients, be able to confidently inform the patients and their family of the patient’s status, and determine future steps for the patient’s therapy regiment. From a broader perspective after speaking with the client Dr. Gogola, we envision that this technology can be expanded to help other causes of muscular disorders such as stroke, arthritis, and sports injuries. The development of a quantitative standard in conjunction with physicians at Shriners Hospital has tremendous implications for the study of upper extremity and muscular dexterity ailments worldwide.
We would like to thank Dr. Eric Richardson, Dr. Gary Woods, Dr. Andrew Dick, Dr. Leroy Chiao, and Dr. Maria Oden at Rice University’s Oshman Engineering Design Kitchen. In addition, we would like to thank Dr. Gloria Gogola at Shriners Hospital in Houston, TX, USA, and Richard and Karen Whitney for their generous funding contribution.