NeckQuest

Inspiration

Students often spend many hours studying, coding, writing, or preparing for exams while sitting in front of a laptop. During these long sessions, neck and shoulder muscles can stay slightly tense without the student noticing it. Usual productivity tools can remind someone to take a break, but they cannot tell whether the body actually recovered.

NeckQuest was inspired by this gap: students need feedback not only about time spent studying, but also about hidden physical overload during study sessions.

What it does

NeckQuest is an AI-powered EMG wearable for students. It measures electrical muscle activity from the upper trapezius, classifies the muscle state as relaxed or tense, and shows whether the student had enough muscle rest gaps during a study session.

The system includes live EMG visualization, guided calibration, machine-learning classification, Holter-style recording, local session storage, playback of recorded sessions, and an interactive browser demo that works without hardware.

The goal is to help students understand when their neck is quietly overloaded during long learning sessions.

How we built it

We built the prototype using an ESP32-S3 board connected to an analog EMG sensor. The firmware samples the EMG signal and supports live streaming and Holter-style recording.

A local Python server handles the web interface, device communication, signal processing, calibration, and machine-learning classification. During calibration, the student records relaxed and tense examples. The server extracts EMG signal features and trains a personal classifier using scikit-learn.

The frontend is built with HTML, CSS, JavaScript, and Canvas. It shows the live signal, calibration state, device status, classification output, and recorded Holter sessions. We also created a hosted GitHub Pages demo so judges can explore the idea without the physical device.

Challenges we ran into

The biggest challenge was working with real EMG data. The signal is sensitive to electrode placement, skin contact, cable movement, ADC noise, WiFi noise, and calibration quality.

Another challenge was making the system useful for students instead of only showing a raw graph. Raw biosignal data is hard to understand, so we needed to convert it into simple feedback: relaxed, tense, and missing rest gaps.

We also had to make the project understandable online, because not every judge will have access to the hardware. That is why we added screenshots, playback, a video demo, and a browser-based demo.

Accomplishments that we're proud of

We are proud that NeckQuest is more than a mockup. It connects wearable hardware, EMG signal processing, personal machine-learning calibration, live visualization, Holter-style recording, playback, documentation, and a hosted demo.

We are also proud that the project uses AI in a practical way. Instead of adding AI as a chatbot wrapper, NeckQuest uses machine learning to interpret noisy biosignal data and turn it into feedback that can help students during real study sessions.

What we learned

We learned that AI for wearable health and productivity tools depends on the whole pipeline: sensor placement, signal quality, preprocessing, calibration, model training, user interface, and trustable feedback.

We also learned that student productivity is not only about focus and time management. Physical fatigue matters too. A student can keep working while their neck muscles stay tense for too long, so detecting recovery gaps can be valuable.

What's next for NeckQuest

Next, we want to add a second EMG channel for the other side of the neck and shoulders. This would allow NeckQuest to compare left and right muscle activity and detect asymmetric tension during laptop use.

We also want to improve the AI layer with better feature extraction, longer-session analytics, personalized study-session reports, and an AI coaching assistant that explains when the student’s body needs a real recovery break.

Long term, NeckQuest could become a student wellness companion for study marathons, exam preparation, online learning, and coding sessions.

Built With

• ai
• arduino
• biosignals
• c++
• emg
• esp32
• health-tech
• javascript
• machine-learning
• python
• scikit-learn
• wearable