NeuroFlex

NeuroFlex addresses the specific problem of inaccessible and costly neuromuscular monitoring particularly with traditional EMG methods, which are clinically oriented, invasive, financially burdensome, and inaccessible. Globally, 14 million people are affected with neuromuscular diseases such as ALS, highlighting the need for accessible diagnosis. Furthermore, a typical on-site EMG costs anywhere from $289 to $496. Finally, 48% of people have to wait over a year in order to be diagnosed -- a crucial time frame as many Neuromuscular diseases take 2-3 years to manifest into their late stage.

NeuroFlex is a cost-effective, wireless, and user-friendly surface Electromyography (EMG) that revolutionizes at-home neuromuscular monitoring. Our bluetooth-based circuitry requires no clunky wires and connects straight to the app. Additionally, our product uses a dry electrode made of Titanium for reusability, and a backing made out of thermoplastic urethane (TPU). Our innovation utilizes bio-compatible polymers such as Thermoplastic Urethane and Polypropylene to encase the technology. Alongside our physical product, we have an app that applies a Convolutional Neural Network to analyze muscle contractions as a continuous time-series for classification. NeuroFlex is unique compared to other portable EMGs in the market because of our Machine Learning model, comfort, and cordless application. With every neuromuscular condition-prone individual utilizing NeuroFlex, we will substantiate the significance of muscle health while delivering an efficient consumer grade product.

Our solution serves those in need of muscle health monitoring, who do not have the means to constantly make the trip to a clinic or purchase an overly priced EMG machine. The older adult demographic is often ignored when it comes to their Neuromuscular health. In their age especially, they are more susceptible to diseases such as ALS. Providing them a chance to be able to regularly check up on their muscle health themselves is crucial to them living a safer and more relaxed life. Additionally, want to better the lives of the 14,000,000 people who currently have a Neuromuscular disease. By providing them with an affordable and accessible way to monitor their muscle health, they can not only feel more at ease, but also have more data to show their doctor. 

Our team is well-acquainted with several ALS patients. We have seen first hand the pain and suffering that afflicts them: constrained to a wheelchair, forced breathing tubes through the nose, and weakened facial muscles. We are looking for a way to truly make a change in the world. We have been inspired by the numerous videos of celebrities participating in the ALS ice bucket challenge, a trend that both ignited our passion and called us to act. Our EMG design was meant to avoid common pitfalls in many wearable devices: awkward attachment to the skin and painful detachment when removing it. We wanted NeuroFlex to be as comfortable for our users as possible. With that in mind, we decided to utilize the TPU substrate. 

Furthermore, we foresee NeuroFlex’s software side as the potential to turn into a database. Data on existing muscular conditions remains limited. However, if more and more people use NeuroFlex, we can amass much data, which can be used for further research and to help propel institutions towards cracking the code on tackling NeuroMuscular diseases.

Currently in the patent-pending phase, we anticipate formal acceptance shortly. While we have yet to commence sales, our progress toward obtaining FDA approval and clarifying regulatory requirements underscores our commitment to compliance and market readiness. This promising venture stands at the threshold of transforming wearable technology and healthcare monitoring. We have a fully-functioning prototype, assembled completely with out key components: the circuit board, the electrode, and the TPU substrate. In addition to the EMG itself, we have a finished app, and our Convolutional Neural Network model has a recall of 96.9% and an accuracy of 98.9%. Finally, we have been able to work with a few ALS patients, and indeed, NeuroFlex was able to discern irregularities in their muscle movement through the EMG readings. 

We believe Solve offers a platform to not only elevate our visibility but also to present our innovative solution on a global stage. Our goal is to make a significant impact by aiding millions of individuals suffering from neuromuscular diseases, and we see Solve as a crucial partner in realizing this ambition. As ambitious high schoolers, we aim to truly change the world.

Our solution is innovative because it is the only surface EMG in the market that utilizes a titanium electrode, a unique approach considering that most others use silver chloride. We avoid using electrode gels as there is the common issue of sticking too hard to the skin, causing irritation when being removed, and its one-time use. Titanium avoids this as its performance never degrades and always provides accurate electrical readings, as well as being reusable. We expect our solution to have an impact as it leverages the harmony of machine learning and medical technology. Our use of a Convolutional Neural Network is unique in that it is the only EMG on the market that is capable of giving both a muscle movement reading and an ALS diagnosis. As a result, we provide an easy method for people to track their health and guide them in understanding what is going on in their body. 

In addition, NeuroFlex is easily scalable and expandable. Currently, NeuroFlex focuses on analyzing and diagnosing, but it can be used for so much more. For example, with more and more users, we can build a giant database, holding a variety of EMG readings and our model’s respective analysis on them. While we can use this to further our own research and development, we can provide research institutions with the data we have collected. Our mission from the beginning was to work towards a brighter future for all patients, and thus, if we can aid the efforts of other innovators and researchers, we will be one step closer to solving the bigger problem. 

Our surface EMG does not require an electrolyte gel to function like a traditional one does. We use dry titanium-film electrodes to increase reliability and promote eco-friendly reusability. The titanium film is on a thermoplastic urethane (TPU) substrate, which is both biocompatible and more biodegradable than other plastics. Titanium is used because it is stronger and more chemically stable than other electrodes. With traditional EMG’s, there is a raw signal amplification that occurs first between each of the signal electrodes and the reference electrode via differential amplifiers. Afterwards, the signals have their highest and lowest frequencies filtered out, leaving the 20-200 Hz range to then digitize and send to the host device. We reversed this process to allow for cordless patches; we can filter out the high and low frequencies on the board, send this signal over bluetooth, then, on the host device’s software, do a digital differential amplifier on all 3 signals. Powered by a small coin battery, the required filtering components and digitizing components are made of simple resistors, capacitors, and amplifiers, and the bluetooth module can be fitted from an already existing chip. Our app is created using the React Native JS Framework with Google’s FirebaseDB as the backend to securely maintain record of the user history and provide authentication. Our app utilizes the Bluetooth Low Energy (BLE) protocol, a power efficient protocol supported by standard mobile operating systems, to connect our EMG with the user's device. This is done using the react-native-ble-plx library which allows connection to the Bluetooth module of the patch. Since the patch operates using BLE, it is possible for the user to keep on the patch for longer periods of time without wasting battery and hindering daily activities for more accurate results. Once the data has been collected, some preliminary data collected as summary statistics and an overall report. The Neural Network (NN) model is a combination of 3 Convolutional layers and traditional 2 Linear layers to allow for the extensive analysis of the data. We used publicly available data collected from 25 patients aged between 21-37 with myopathy and ALS to train our NN model. With this structure, our NN has been able to achieve a recall of 96.9%.

Our team is founded on the idea that no matter who you are or where you are from, as long as you are dedicated to our mission of saving people’s lives, we welcome you to join the fight. We do not tolerate any form of discrimination, harassment, sexism, racism, or hate speech.  Furthermore, we promote diversity and women in STEM, especially encouraging members of the LGBTQ community to have access to entrepreneurship opportunities. One of our team members knows the struggles of being lesbian women in STEM. She has undergone fierce prejudice and been looked down upon for her identity, however still strives to provide resources and accessible medical diagnosis for the rest of the world. Through her experiences, our team sees the struggles that LGBTQ individuals go through. Our staff also features a diverse group, with our members being Mexican, White, Chinese, and Indian, who have experienced the struggles of being minorities. To provide opportunities to all, we provide resources and educational talks, and openly encourage all individuals to pursue their passions in STEM.