The continuing evolution of orthopaedic solutions to common foot and ankle problems means engineers and clinicians need increasingly complex and physiologically correct models of human movement. The design uses new parts and improved methods designed to interface with current machines and operating procedures that simulate the movement of human walking. The team added three additional tendons to the foot-ankle model and replaced the current tendon actuators with more robust models, capable of larger forces and faster movement. These additional tendons resulted in the design, validation and implementation of additional actuators, load cells and accompanying electronics. Adding these components required designing a spacing flange to maintain physiologically accurate tendon pull direction while providing the necessary space for the movement systems. The team overhauled the electrical system so it could provide necessary power to all components. The sponsor can use these new additions to the robotic gait simulator to better model orthopaedic solutions to gait defects, such as poor posture and injury.

The sponsor manufactures the EMILY lifeboats used remotely in various lifesaving rescue missions across the globe. The remote controllers used to operate the EMILY buoys are no longer in production, so the team designed and fabricated a new controller that meets the performance benchmarks of previously used controllers while adding some new functionality.Operating within the 2.4 GHz industrial, scientific and medical band, the controller sends control signals to EMILY and receives battery telemetry back. The control channel sends throttle, rudder direction and engine polarity controls to the EMILY. The controller is powered by an AA battery and can operate at a distance of 1 mile over calm water while maintaining a strong battery life of two hours. The controller has a waterproof rating of International Protection, or IP, 67, meaning that the device can be submerged in water for up to five minutes at a depth of 1 meter.

Medical imaging is a critical component of disease diagnosis in medical practice. Although health care providers have used advances in medical imaging technologies to implement more effective point-of-care strategies, the analysis of medical images remains inefficient and highly subjective. The solution developed is a full image classification, feature extraction and feature analysis tool called FractalEyes. The system can classify an input image, perform application-specific feature extraction and analyze the features to provide health care providers with a more quantitative measure of diagnosis. The software uses a mutual information classification algorithm from the scikit-learn Python library. The program communicates both graphically and quantitatively which features of the image are useful in the image’s classification. The team used this proof-of-concept system to classify histological images of human cells. The idea can be extended to applications ranging from disease diagnosis to counterfeit item detection.

Sounds associated with health and illness are vital for health care diagnoses and therapy. This project focuses on developing a system capable of processing and extracting additional features from sound and speech during a health care encounter. This recording and analysis may help with the treatment of a variety of diseases.The sound capture system uses high-fidelity microphones positioned around the room to record, store, analyze and display sound components of speech and other sound-generating aspects associated with the physical examination, such as breath sounds. The system also includes speech-to-text analysis to produce a transcript of the encounter. The system incorporates natural language processing to analyze conversations and organize useful information by helping with documentation of the patient’s condition. The final product provides the user with valuable data that may be incorporated into the electronic medical record system within the context of a digital “wired room.” This system can improve efficiency in medical care, improve diagnoses and reduce the time a physician needs to spend with each patient.

Approximately 30 million individuals in the United States suffer from eating disorders. Current treatments include physiological therapy and nutrition education. The system designed is a novel treatment approach that contains a progressive, self-directed software exposure experience to “desensitize” and acclimate individuals suffering from anorexia.The software exposes users to different scenarios through a virtual reality program created in Unity and displayed on the Oculus Go, a head-mounted VR display. The software contains nine modules that increase in difficulty while incorporating various scenarios with food, people and environments. Analog biosensors and an Arduino processor record skin impedance, pulse rate and respiratory rate to characterize the user’s stress response. The physiological data collected is displayed in real time in the Oculus Go to enhance the user’s understanding of their physical response to the virtual reality environments.

More than 350,000 cardiac arrests occur outside of hospitals in the United States annually. Many people routinely perform CPR to increase blood flow and deliver oxygen, but it must be performed correctly to be effective. The system designed immerses the user into a CPR scenario in virtual reality, or VR, using Leap Motion hand-tracking technology and an HTC Vive tracker to integrate a CPR mannequin into VR. The user wears a glove containing an accelerometer, programmed with an Arduino Micro worn on an armband, to detect the depth, frequency and acceleration of their compressions on the mannequin. For more effective CPR training, the user is provided with real-time auditory and visual feedback to correct CPR compressions. Speech recognition programming allows the user to communicate with a non-playable character in the scenario, who will recognize key phrases such as “call 911.” The system provides the user with a comprehensive score at the completion of a training session to track their improvement as they do additional training in accordance with the American Heart Association’s guidelines.

Medications and implantable devices can adversely affect the mechanical properties of platelets when physical stresses are introduced. Applied stresses, such as shear forces, pressure and acoustic-mediated vibrations, can activate the circulating cells driving blood clot formation, which may lead to ischemia, heart attack and stroke. By measuring cell stiffness, the likelihood that cells will activate under applied stresses can be determined. The designed system induces the electrodeformation of platelets by dielectrophoresis, which uses a nonuniform electric field to trap and measure cells’ mechanical properties, such as stiffness, without causing activation. As the voltage is varied and the cells are deformed, the system images and tracks the parameter changes that occur in each of the visible platelets. The measured changes in platelet parameters are sent to the system’s programmed algorithms to
calculate the mechanical properties of each individual cell. The images and cell properties can be monitored through a user-friendly graphical user interface.

Image capture is a very sensitive process. Any noise induced during the process lowers the quality of the image captured. A common type of noise is electrical noise from the power supply. The sponsor asked the team to determine how different power supply configurations can affect image quality of a complementary metal oxide semiconductor image sensor. To perform the tests, the team designed a camera system with a modular power supply to allow easy switching between different power supply configurations. Test configurations designed included linear, switch-mode, and a combination of linear and switch-mode power supplies. The team qualitatively assessed the captured images with each configuration and quantitatively calculated the signal to noise ratio, or SNR, with Matlab. The image with the highest SNR was considered the highest quality image. The power supply with the highest quality image was recommended for further use in any image-capture applications. The test configuration with the highest presence of noise should be investigated further to more thoroughly understand the root cause of the noise.

The sponsor wants to assemble an electric vehicle using commonly sourced components from salvage vehicles. These salvage parts, though functional, require extensive work to integrate due to their proprietary controller area network, or CAN, bus systems. Correct CAN bus codes are required to operate these components in the system. The design approach for this project required integration and assembly of these components into one cohesive electric vehicle system that can be controlled by an operator.
The team created new software and modified existing software to read, interpret and forward CAN codes throughout the components of the system. The system incorporates micro-controllers that work in conjunction with the CAN codes and receive input from the operator. The team developed software and hardware to control and display voltage, temperature and power-sensing systems. The system is fitted into an existing frame and configured to run a cooling system that regulates temperatures across sensitive components based on logic created by the team.

The autonomous driving system designed can make steering calculations using a proportional-integral-derivative controller and a servo motor. The system uses a GPS to detect its position in real time. These components, in combination with a lidar and linear actuator, allow the system to detect objects in its front view and stop accordingly to avoid collisions. The user is also able to deactivate and turn off the system with a kill switch. An Arduino Mega microcontroller controls the system. The kit manufactured allows a solar-powered go-kart to drive autonomously. There are plans to integrate this kit into the Racing the Sun competition.