Capstone 2021 Biomedical Engineering

Alexander Horn, Seth Lombardy, Alexi Starks
Faculty Advisor: Uri Feldman

Pediatric wheelchairs on the market today are often overpriced for the short period of time they are used due to the continuously growing patient. The baseline for these wheelchairs is security and safety of the patient while allowing full control over mobility within the limits of the powered wheelchair. With these standards, there is an expectation to meet a certain degree of reliability, affordability, durability, and efficiency. To meet all these factors often high-quality material will be compromised as a result. For example, the use of PVC framing to maintain affordability while losing durability (compared to common materials). The A.M.P wheelchair has components that will maintain the integrity of day-to-day use without the prohibitive cost of buying a commercial wheelchair. Complete assembly of the wheelchair only includes a limited amount parts, allowing for easy assembly and modification as the child grows. Both cheap and quick, interchangeable PVC piping allows for the chair to meet any size criteria (within integrity of piping) ensuring that this chair can be used year after year. Another feature includes the Arduino driven motor controller that allows for the patient to control their movement through a joystick interface. This specific control is design for 360-degree movement with quick and easy adaptability for the patient. The prediction of reliability was given through assumption of components full potentials, pre-testing statement will be concluded with sections of evidence.

View the PDF for A.M.P. Wheelchair

Nicole Cobuccio, Rachel Kelleher, Anthony Maida
Faculty Advisor: Ata Kiapour

Pill bottles are designed to safely store medicine, as well as to keep them out of the reach of children. They are very effective in keeping the medicine safe; using different mechanisms to effectively lock the bottles. These mechanisms often require a lot of hand strength to open. However, many older adults deal with muscular or neurological disorders that can make it difficult to bypass the childproof locks. To address this problem, we have designed a universal cap which will dispense the pills with minimum force while remaining childproof to keep the medication safe. The Carpal Cap will dispense pills using a lever activated by a magnetic dongle; removing any frustration for adults with muscular or neurological disorders that prevent them from opening a regular pill bottle, and creating enough threshold to not allow a child to retrieve any medication. Over the course of two semesters, we have tested different designs and iteratively improved our concept. The current prototypes are designed to fit the common pill bottles. We have used an anonymous survey to validate the unmet clinical need and our concept.

View the PDF for Carpal Cap - Easy Pill Dispensing

Michael Callahan
Faculty Advisor: Uri Feldman

The Shoulder Joint assessment system is a contactless way to determine human body joint angles. The input is an under two minute exercise video where the patient moves their arm through six motions, abduction, adduction, internal rotation, external rotation, forward flexion, and extension. These six measurements define the range of motion (ROM) for the shoulder joint. In the conventional setting a physical therapist would guide their patients through the motions, placing a goniometer to measure each. The Shoulder Joint assessment system uses an operator to look through the exercise video for six frames, each capturing the maximum displacement for one of the six motions. Past identification of the six frames the system is fully autonomous and takes advantage of OpenPose. OpenPose is a machine learning model that recognizes humans in an image and predicts their pose, or joint locations like the wrist or shoulder. These joint locations are (X,Y) pairs that detail the location in the image. From two points the angle from horizontal can be calculated. Abduction, adduction, forward flexion, and extension angle from horizontal are determined using the shoulder as the origin and the wrist as the extreme point. Internal and external rotation are determined using the elbow and the wrist. Finally, the angle from horizontal is converted to clinically accepted ranges, i.e. abduction is converted to angle from vertical. These 6 angle measurements in clinically accepted form are exported as a JSON so an EHR may easily read it in.

View the PDF for Contactless Shoulder Range of Motion Assessment System

Jarod Tessier, Sage Williams
Faculty Advisor: Weihui Li

'The main goal of this project was to create a system for hospitals’ medical equipment that tracks and locates the devices through a database. Querying the database would enable hospital staff to quickly and efficiently locate devices for use throughout any given workday. RFID technology is the most effective method for tracking, and it was the chosen method for this project. Sensors would be applied to various key doorframes on a hospital’s floor allowing for any devices moving through to be tracked. The RFID tags are able to pickup the transmitted signals from the sensors and relay them back, enabling the whole system to function properly. Arduino sensors were used for the testing purposes of this project, but higher frequency sensors would be most effective in the relaying of signals to the tags.

View the PDF for Hospital Device Tracking System

Riley Bell, Austin Kramer, Mark Offermann, Divesh Patel
Faculty Advisor: Ata Kiapour

'Although post-surgical fatalities are rare, infections of surgical sites are responsible for 77% of these deaths. Post-surgical, long-term wound care is vital, and monitoring the potential for wound infection is an aspect of medicine that requires further development. Research has shown that temperature is potentially the most important metric that can predict surgical site infections. Humidity is another metric that can dictate the rate of proliferation and tissue healing at the wound site. The combination of these two measurements gives potential for a platform to monitor wound infection/healing through a wound dressing with sensor integration. Currently, no streamlined wound infection detecting device exists that is widely adopted. Nurses must routinely change dressings on wounds and detect infections with visual examinations and biopsies. The device developed by our team would ease the burden on medical staff, as well as provide numerical data to accurately detect infectious conditions. We have designed and developed a proof-of concept prototype of an infection detecting system that does not require direct skin contact. This “infector detector” will consist of a standard medical gauze layer with a sensor-web layer on top of it, and a removable, reusable circuit layer on top of that. The sensor layer on top of the gauze will capture ambient temperature and humidity of the wound through the gauze, and relay that information up to the circuit layer. From the circuit layer, depending on the state of the wound environment, a visual signal (LED) will alarm the user of potential infection danger based real time wound monitoring data. The design would fill the need for a wound monitoring system that can generate signals to notify the user or medical personnel of potentially infectious wound conditions using actual temperature data, without the wound having to be exposed and visually examined as the current process demands.

View the PDF for Infector Detector: A Non-Invasive Approach to Track Wound Infection

Daniela Ojeda, Richardson Laforest, Cormac Paterson
Faculty Advisor: Ata Kiapour

The knee is a common point of injury for many aspiring athletes. One group of typical knee injuries may be referred to with the umbrella term, runner’s knee. These are injuries related to overtraining, imbalances, and poor acclimation to the exercise of running or jogging. Running is a standard form of cardiovascular training in both athletes and the general populace and running at any level or style can result in acute or long-term knee injuries based on a set of factors. Some of those factors are not controllable, while others are hard to gauge or are very dependent on the individual. Proper coaching can help athletes avoid many of the hurdles which can lead to injury, but most casual runners do not have the luxury of consistent observation by professionals. The aim of this project was to develop a device which can track the motions of the lower limbs and guide users based on deviations in symmetry and form that may lead to injury down the line. This would lower the barrier for entry for casual runners to train without fear of injury or need to pay for expensive coaching or analysis. Our wearable device uses MEMS technology to track knee motion during daily activities and sporting events. The device will not catch all potential risk factors, but by taking user data from exercise and anatomical inputs, it may target more than one source of risk and assist in its avoidance. We have validated our approach using an anonymous survey and interviews with professional physical therapists and trainers. We have also validated the ability of our device to measure knee angles compared to standard goniometers.

View the PDF for Knee Tracker: A Wearable Device for Knee Motion Tracking

Kirshan Chauhan
Faculty Advisor: Uri Feldman

The motivation for this project is the lack of personalized prognosis of Multiple Myeloma, a type of cancer that causes mutations in patient plasma cells. The current generalized prognosis of Multiple Myeloma is not considering the patient’s genomic composition. Patients are facing far more complications such as, drug resistance and drug related side effects because of the use of generalized prognosis theory. To address this problem of generalized treatment, it is postulated that a computational model could be developed to incorporate the patient’s genomic computation. By developing and providing a computational model, we will be able to create a model that is based on the patient’s genomic compositions in order to fill the void of personalized prognosis. Computational models such as DeepDR models have been developed and have proven to be effective in providing patient specific responses to possible therapeutic agents. (Chiu). Thus, it seems like the creation of a computation model like the models used for Leukemia is possible. In certain Multiple Myeloma subgroups there are also certain drugs known to effectively work for certain conditions. However, personalizing those models to the patient population would provide various benefits ranging from increased drug efficacy to reduced drug-related side effects. Assurance that the model will function as defined can be confirmed through the cross-validation of the model.

View the PDF for Multiple Myeloma Computational Analysis (MMCA) Software

Zack Solovay
Faculty Advisor: Uri Feldman

The device is a Personal air quality sensor, consisting of an air sensor, and Arduino, a power supply, and a display. Often people can suffer from poor air quality without realizing it. Many toxic chemicals are odorless and colorless, and people can expose themselves to harmful levels of a gas without realizing it is affecting them. This can range from highly toxic gasses like carbon monoxide, to even low concentrations of CO2 lowering cognitive function. Poorly ventilated areas can build up toxic gasses and lead to black outs or death of workers. CO2 and other gasses can sink to the floor and displace oxygen or other lighter gasses over time, eventually raising the toxicity of a room or chamber to dangerous levels. The Device displays the ambient CO2 levels and tracks air quality over time, displaying it with a colored led ring. The sensor can detect for a few other gasses, like hydrogen gasses, and other sensors can be swapped out to test for other gasses. Currently the device uses two large LED rings as displays, for clarity and nighttime or underground visibility, but other form factors of display such as LCD or OLED could be swapped out as well. The device was built with modularity in mind, to fit the form factor or specific needs of the user while still serving the same basic function: detection of hazardous chemicals. The goal of the Gas Sensor is to lower the number of deaths from airborne toxicants. Because airborne chemicals may be odorless, colorless, and build up in poorly ventilated areas, many people may be exposing themselves unknowingly to harmful chemicals. This can be detrimental to health, cognitive function, cause blackouts, or death. The Gas Sensor should allow users to detect invisible hazards before incurring harmful effects and allow civilians and rescue workers to properly equip for or avoid a dangerous area. For civilian or everyday use, it will track the CO2 level over time, giving the user awareness into how much the gas could be affecting their lives.

View the PDF for Personal Toxic Gas Detector

John McAreavey, Gabriel Roy
Faculty Advisor: Weihui Li

Infusion pumps are medical devices that are used to deliver medications in syringes via motor powered plunges over a desired period-of-time via intravenous lines. Syringe pumps, and infusion pumps alike, have been plagued with poorly designed user interfaces. Clinicians have reported confusion when inputting data into the devices, monitoring the status of the infusion, and when trying to manage the alarms. Some of these problems include sensors not picking up on potential harm-inducing scenarios, like air-in-line or occlusion. The proposed project will take these problems and fix them to the best of our abilities. The user interface will be redesigned to have a more user-friendly interface with a separate directional keys and number pad. The user-interface will also be able to move around as well to make it easier to see the clear LED display. Along with the redesigned interface will come an updated notification system that makes it clearer as to what is wrong with the syringe pump, when alarms have been tripped, and displays the infusion status, when running normally. Both visual and audio signals will be used to make sure that there is little to no reason that the operator won’t know that the pump is working. The problems with the sensors will be fixed through lowering the tolerances for both air-in-line and occlusion pressure to ensure that the patient is as safe as possible. The goal by the end of the project is to make as cost-efficient as a syringe pump as possible while updating the user-interface and fixing the problems with the alarms.

View the PDF for Pump Up The Jam

Maria Dam, Sean Spencer
Faculty Advisor: Uri Feldman

This project aims to solve an issue encountered in the teaching of medical students in an operating theatre. It is a frequent occurrence that medical students will enter an operating room to observe a procedure as it occurs. This provides a necessary perspective when preparing students to perform their duties as medical professionals to ethical standards of care. However, problems arise because these classes are crowding into the operating room during the procedure. This leaves the surgeon(s) performing the procedure with less room to maneuver and is an added distraction for the working team. There is also the risk of contamination because of additional people in the operating room. We will address this problem with a video system designed to allow maximum immersion within the operating room without the negative consequences of having in-person observation.Our secondary goal is to provide a system that will allow students to observe surgical procedures completely remotely. The COVID-19 pandemic has demonstrated the need for new and enhanced teaching tools. With our device we aim to reduce the strain on educators while providing alternative learning modalities. Our design consists of a suspended-arm mechanism with an attached camera and tracking mechanism which will provide 360° coverage of the operating theatre with the ability to record and stream audio & visual elements. This device is called the Surgical Theatre Teaching Aid (STTA). We anticipate that this project will enhance learning outcomes for lab courses at the Tufts Cummings School of Veterinary Medicine who has been our generous patron from the beginning of the undertaking.

View the PDF for Tele-presence Surgical Theatre Teaching Aid (TSTTA) for Veterinary Surgeons

Devin Wittenberg, Nick Newman
Faculty Advisor: Weihui Li

Amongst the world population, 46% of individuals over the age of 60 possess a form of disability, many of which require the regular use of medication. The MedAssist Device is intended to aid those with disabilities, or elderly patients with taking their medications on time and in the proper quantities. It is all too common for individuals with memory problems to forget to take their daily medications, which could have lasting effects on their health and quality of life. Our device is designed to dispense the proper amount of medication, at the proper times throughout the day to ensure proper intake and dosage.

View the PDF for The MedAssist: Automatic Pill Dispensing Unit for Elderly and Disabled

Anna Wiatrowski, Olivia Rourke, Alexa Cormier
Faculty Advisor: Weihui Li

Being a patient in a hospital can be an overwhelming experience. The vital signs of the patient need to be recorded in order to monitor their health during their time in the hospital. This process often requires the patient to be connected to bulky machines and an entanglement of wires. The Vital Gown allows comfortability and a wider range of motion for a patient that needs their vital signs measured. These vital signs include the use of an ECG and a heart rate monitor. IOT capabilities send the signals measured from the sensors on the patient to a mobile device approximately 15ft away. All of the wires that are still needed to connect from the sensor leads and circuit board are concealed within the fabric of the patient’s hospital gown. The wires are all concealed properly to ensure a proper connection and a strong signal. Both the wireless IOT connection and the concealment of the wires allows the patient to feel more comfortable and like less of a lab rat while their vital signs are being recorded. They can move freely in their hospital beds and even in their hospital rooms while the signals are being sent. In the future, the Vital Gown design can be expanded to measure additional vital signs, a longer distance IOT connection, and a more comfortable and appealing gown.

View the PDF for The Vital Gown

Roi Bubjaku, Kennedy Nguyen, Nihal Reddy
Faculty Advisor: Uri Feldman

This project is intended to design and manufacture a form of hand and wrist rehabilitation therapy for post-surgery patients. [1] Many factors contribute to hand/ wrist surgery such a carpal tunnel release which is responsible for over 400,000 procedures in the US yearly. Physical therapy is required following a procedure as it helps a person regain control, balance and range of motion while increasing the overall healing processes. Our goal is to create a more engaging form of physical therapy through repeated motions using the device on videogames specifically made for the purpose of rehabilitation. The device will function as a 3-D space mouse which will use sensors to pick up movements from the hand/ wrist to use as the input for the movements of the cursor on the screen. The sensors would have to function at specific angles which are measured from current physical therapy exercises so that the user receives proper range of motion and strengthening training while being able to surf the web at their leisure. There were few devices which used therapy motions as tracking for use of cursor movements of a mouse. Even more the device can be intended to be used for entertainment purposes. This device is even more specific by isolating strictly wrist movements. The project is feasible for our group to have a working prototype as there are multiple tutorials online on how to create an air-controlled mouse which we will combine with the parameters required for physical therapy exercises which are also thoroughly detailed online.

View the PDF for Wrist Therapy System

Ali Sirry, Khalid Al-Mharib, Talal AlWaraq
Faculty Advisor: Allen Kiapour

Biomedical engineering has made it easy for people to diagnose symptoms that cannot be physically diagnosed. Pain is a physical sensation that is hard to define and diagnose, especially when patients are unable to express themselves. Over a million people die each year from a car accident and more than 20 million people could have a major disability. These statistics could drastically decrease if the world has a device that could detect the exact spot of a pain in the body or a nerve. All the individuals suffering from a fracture, rupture, or inflammation, would have to go to the hospital, and by the time they get diagnosed they could experience more pain and could develop other unwanted symptoms. The time taken to transfer the patient from the accident location, to the ambulance, and then to the MRI device could easily be reduced into just a single loop. That is the Pain Signal Detector which could be used anywhere, and have all the information about the abnormal pain signals be delivered to their primary doctor before admitting the patient into the hospital, and from there the doctor could decide whether it is an emergency or not. PSD is a revolutionary device that could take the place of a lot of medical imaging devices, because it is accurate and precise when it comes to the intensity of the emotional experience.

View the PDF for Pan Signal Detector Device

Justin Chim, Ahmed Tolah, Dylan Lindsay
Faculty Advisor: Allen Kiapour

A major issue for many aging individuals and those with certain ailments are muscle imbalances, muscle deterioration, and nerve or strength degradation. Our group wanted to design a project that helps clinicians analyze data from different types of people and especially design devices to aid each person individually. Current methods to help people with balance problems is physical therapy or expensive motion tracking. There currently is no middle ground to give an affordable alternative for people that require a small amount of dianositics.

View the PDF for Fall Analysis Device

Matt Clark, Matt Raples, Saad El Hassouni
Faculty Advisor: Allen Kiapour

Maximizing training efficiency is an ongoing problem in the fitness industry. Companies like Whoop and Biostrap have been constantly working towards improving performance amongst athletes by using biometrics to optimize muscle performance, recovery, and training. Specifically, the use of electromyography (EMG) is a measurement that these companies, are trying to incorporate into their fitness devices, so that they can more accurately predict recovery time and training for their users [3]. The reason why it is difficult for these companies to implement EMG into the device is because EMG signals are recorded at the muscle corresponding to the electrode placement. With this in mind, the goal is to determine if EMG signal can be used to calculate the 1RM of a Deadlift (DL) for athletes using Rate of Perceived Exertion Training Programs.

View the PDF for Myo-Efficiency

Noah Niles, Kevin Devine, Jacob Girard
Faculty Advisor: Allen Kiapour

Wireless power transfer (WPT) describes the transfer of energy from a source to a load while travelling across a medium. Since WPT does not require direct contact, it is a promising means of powering and controlling implantable medical devices of the future. The time frame for current internal batteries is generally between 7-12 years. At the conclusion of the batteries life, the replacement process will consist of a surgical procedure which increases the risk for the patient. In our research and design, we explore the dynamic levels of WPT to converge at the final ability to charge a device within the body externally.

View the PDF for Wireless Power Transfer for Implantable Medical Devices

Haley Dionne, Colleen Walker
Faculty Advisor: Allen Kiapour

The CalmDown Bracelet is designed to aid people who suffer from any range of anxiety and panic attacks, large or small. This device is designed to monitor the patient's heart rate, sense when their heart rate is rising in “panic attack” mode and help bring their heart-rate back to resting through guided meditation breaths.

View the PDF for The CalmDown Bracelet

Molly Donahue, Alexandria Rodriguez
Faculty Advisor: Allen Kiapour

The goal of this project is to solve an ongoing medical phenomenon known as implant migration, related to the birth control implant. As two women who use contraceptives and are aware of this issue, we chose this project to investigate methods to prevent this serious medical complications unrelated to the hormones or implant itself.

View the PDF for Flexplanon: The Solution to a Stable Arm Implant